BEE Exam Prep › Paper-1 › Chapter 9

BEE Paper-1 — Chapter 9: Energy Monitoring & Targeting

116 questions — 56 objective (1 mark), 44 short (5 marks), 16 long (10 marks). Every answer is checked against the 2014 BEE guidebook and carries its book section reference plus an explanation.
▶ Practice this chapter interactively (timer, read-aloud, progress saving).

Objective questions (1 mark) — 56

📖 §9.6 Linear Regression — E = C + M·P

1. A factory has a fixed energy consumption of 2,000 kWh/month and it consumes a total of 38,000 kWh/month for manufacturing 90,000 units of the product. The variable energy consumption in kWh/unit is ____.

  1. 0.4
  2. 2.4
  3. 2.5
  4. none of the above
Answer: A) 0.4
Confirmed vs Book-1 §9.6 — the standard performance line is E = M·P + C, so the variable (slope) term M = (E - C)/P. M = (38,000 - 2,000)/90,000 = 36,000/90,000 = 0.4 kWh/unit. C = 2,000 kWh/month is the fixed/base-load part and is excluded before dividing by output. Answer (a).
Source: Sep 2021
📖 §9.6 Plant Energy Performance (PEP) & production factor

2. Which of the following data is not used for calculating Plant Energy Performance?

  1. Reference year energy use
  2. Production factor
  3. Current year energy use
  4. Maximum electrical demand
Answer: D) Maximum electrical demand
Confirmed vs Book-1 §9.6 (M&T normalisation) — PEP% = (Reference-year equivalent energy - Current-year energy)/Reference-year equivalent × 100, where Reference-year equivalent = Reference-year energy × Production factor and Production factor = Current-year output / Reference-year output. Only reference-year energy, production factor and current-year energy are needed; maximum electrical demand (kVA) plays no part. Answer (d).
Source: Sep 2021
📖 §9.6 Specific Energy Consumption (Figs 9.7-9.8)

3. The ratio of energy consumption to corresponding production quantity is called

  1. energy performance
  2. specific energy consumption
  3. production factor
  4. specific production ratio
Answer: B) specific energy consumption
Confirmed vs Book-1 §9.6 — Specific Energy Consumption (SEC) = energy consumed / corresponding production (e.g. kWh/tonne, toe/tonne). 'Production factor' is current output ÷ reference output, and 'energy performance' is the % improvement figure - neither is energy per unit output. Answer (b).
Source: Apr 2010
📖 §9.6 CUSUM Charts (Fig 9.12)

4. A CUSUM graph follows a random fluctuation trend and oscillates around

  1. 50% line
  2. 100% line
  3. 0 line
  4. mean value line
Answer: C) 0 line
Confirmed vs Book-1 §9.6 — 'A typical CUSUM graph follows a trend and shows random fluctuation of energy consumption and oscillation around zero (baseline or standard).' CUSUM = running Σ(E_actual - E_calculated); when only random variation exists the positive and negative differences cancel, so the plot hovers about the 0 line. Answer (c).
Source: Apr 2010
📖 §9.4 Benefits of M&T

5. The goal of using energy monitoring and targeting (M&T) is to

  1. determine the relationship of energy use to key performance indicators such as production, rejects etc.
  2. identify and explain increase in energy cost
  3. draw energy consumption trends (weekly, seasonal, operational)
  4. accomplish all the above
Answer: D) accomplish all the above
Confirmed vs Book-1 §9.4 (Benefits of M&T) — the listed benefits include relating energy use to key output/performance indicators, identifying and explaining an increase or decrease in energy use, and drawing energy consumption trends (weekly, seasonal, operational). All three are stated goals, so 'all the above'. Answer (d).
Source: Apr 2010
📖 §9.6 Linear Regression — E = C + M·P

6. The fixed energy consumption of a company is 1000 kWh per month. The line slope of the energy (y) versus production (x) chart is 0.3. The energy consumed in kWh per month for a production level of 80,000 tons/month is

  1. 24,000
  2. 24,100
  3. 25,000
  4. 38,000
Answer: C) 25,000
Corrected (was b) — Book-1 §9.6: E = M·P + C with M = slope = 0.3 and C = fixed = 1,000 kWh/month. E = 0.3 × 80,000 + 1,000 = 24,000 + 1,000 = 25,000 kWh/month. Option (b) 24,100 wrongly adds 100 instead of the 1,000 kWh fixed load; the variable part alone (24,000) is option (a). Answer (c) 25,000.
Source: Apr 2010
📖 §9.6 Specific Energy Consumption — units

7. Which of the following terms does not refer to specific energy consumption?

  1. kWh/ton
  2. kCal/ton
  3. kJ/kg
  4. kg/kCal
Answer: D) kg/kCal
Confirmed vs Book-1 §9.6 — SEC is always ENERGY per unit of OUTPUT: kWh/ton, kCal/ton and kJ/kg are all valid SEC units. kg/kCal is output per unit energy (the inverse), so it is not a specific energy consumption unit. Answer (d).
Source: Apr 2010
📖 §9.6 Normalisation of data

8. The process of removing the impact of various factors on energy use so that energy performance of facilities and operations can be compared is called

  1. Averaging
  2. Normalization
  3. Tracking
  4. Optimization
Answer: B) Normalization
Confirmed vs Book-1 §9.6 — removing the influence of variable factors (production level, weather/degree-days, occupancy, product mix) so that facilities and periods can be fairly compared is called normalisation. Regression/production-factor methods in M&T are the normalising tools. Answer (b).
Source: Apr 2010
📖 §9.6 Plant Energy Performance (PEP) & production factor

9. For calculating plant energy performance which of the following data is not required

  1. current year's production
  2. reference year's production
  3. reference year energy use
  4. capacity utilization
Answer: D) capacity utilization
Confirmed vs Book-1 §9.6 — PEP needs: reference-year energy use, reference-year production and current-year production (to get the production factor), and current-year energy use. Capacity utilisation is not part of the calculation. Answer (d).
Source: Nov 2009
📖 §9.6 CUSUM Charts

10. In the first two months the cumulative sum is 4 and 12 respectively. In each of the next two months Ecalculated is more than Eactual by 3. The energy savings at end of the fourth month would be

  1. -6
  2. 0
  3. 6
  4. none of the above
Answer: C) 6
Confirmed vs Book-1 §9.6 — CUSUM = running Σ(E_act - E_calc). After two months CUSUM = +12. If E_calc exceeds E_act by 3 in each of the next two months, each difference is -3: CUSUM = 12 - 3 = 9, then 9 - 3 = 6. The line FALLS from 12 to 6, and the size of the fall (12 - 6) = 6 is the energy saved in those two months. Answer (c) 6.
Source: Nov 2009
📖 §9.6 Linear Regression — E = C + M·P

11. In an industry the average electricity consumption is 4.6 lakh kWh, average production 40000 tons with specific electricity consumption of 10 kWh/ton. The fixed electricity consumption for the plant is:

  1. 60000 kWh
  2. 46000 kWh
  3. 20000 kWh
  4. none of the above
Answer: A) 60000 kWh
Confirmed vs Book-1 §9.6 — variable (production-related) energy = specific consumption × production = 10 × 40,000 = 4,00,000 kWh. Fixed C = total - variable = 4,60,000 - 4,00,000 = 60,000 kWh. C is the y-intercept of the energy-vs-production line. Answer (a).
Source: Nov 2009
📖 §9.6 Linear Regression — E = C + M·P

12. In an industry the average electricity consumption is 10 lakh kWh for a given period. The average production is 90,000 tons with a specific electricity of 10 kWh/ton for the same period. The fixed electricity consumption for the plant is

  1. 1,00,000 kWh
  2. 9,90,000 kWh
  3. 10,000 kWh
  4. none of the above
Answer: A) 1,00,000 kWh
Confirmed vs Book-1 §9.6 — variable energy = 10 kWh/ton × 90,000 tons = 9,00,000 kWh. Fixed C = 10,00,000 - 9,00,000 = 1,00,000 kWh, i.e. the base load that persists at zero production. Answer (a).
Source: 2019
📖 §9.6 XY Scatter Diagram (Fig 9.10)

13. Large scattering on production versus energy consumption trend line indicates

  1. Poor process control
  2. Inefficient equipment
  3. Inefficient process
  4. None of the above
Answer: A) Poor process control
Confirmed vs Book-1 §9.6 — 'If data fit is poor, it indicates poor level of control and hence a scope for energy savings.' Large scatter about the production-vs-energy trend line therefore signals poor process control (and high savings potential), not necessarily inefficient equipment. Answer (a).
Source: 2018
📖 §9.6 Linear Regression — E = C + M·P

14. If the fixed energy consumption of a company is 2000 kWh per month and the line slope of the energy (y) versus production (x) chart is 0.3, then the energy consumed in kWh per month for a production level of 60,000 tons/month is _______.

  1. 16,000 KWh
  2. 18,000 KWh
  3. 22,000 KWh
  4. none of the above
Answer: D) none of the above
Confirmed vs Book-1 §9.6 — E = 0.3 × 60,000 + 2,000 = 18,000 + 2,000 = 20,000 kWh/month. 20,000 kWh is not among options (a)-(c) (18,000 is the variable part only), so the correct choice is 'none of the above'. Answer (d).
Source: 2018
📖 §9.6 Linear Regression Analysis

15. ______ is a statistical technique which determines and quantifies the relationship between variables and enables standard equations to be established for energy consumption.

  1. linear regression analysis
  2. time-dependent energy analysis
  3. moving annual total
  4. CUSUM
Answer: A) linear regression analysis
Confirmed vs Book-1 §9.6 — 'Linear regression analysis is a statistical technique which determines and quantifies the relationship between variables... enables standard equations to be established for energy consumption.' CUSUM plots cumulative differences, MAT sums 12-month totals and time-dependent analysis only plots energy against time. Answer (a).
Source: 2017
📖 §9.6 Linear Regression — E = C + M·P

16. In a manufacturing plant, following data are gathered for a given month: Production - 1200 pieces; specific energy consumption - 1000 kWh/piece; variable energy consumption - 950 kWh/piece. The fixed energy consumption of the plant for the month is -------

  1. 6,000 kWh
  2. 10,000 kWh
  3. 12,000 kWh
  4. 60,000 kWh
Answer: D) 60,000 kWh
Confirmed vs Book-1 §9.6 — fixed share per piece = SEC - variable = 1000 - 950 = 50 kWh/piece. Fixed energy for the month C = 50 × 1200 pieces = 60,000 kWh (total 12,00,000 kWh minus variable 11,40,000 kWh). Answer (d).
Source: 2017
📖 Book-1 PAT / M&V (outside Ch-9 §9.1-9.7)

17. M & V audit under PAT is carried out

  1. Immediately after the baseline audit
  2. Every year following the baseline audit
  3. At the end of each PAT cycle
  4. Before the baseline audit
Answer: C) At the end of each PAT cycle
Confirmed — Book-1 PAT/M&V (outside the Ch-9 text): under Perform-Achieve-Trade a baseline (energy) audit fixes the reference year, and the Monitoring & Verification (M&V) audit by an empanelled verifier is carried out at the END of each PAT cycle to verify the SEC reduction achieved against target. Answer (c).
Source: 2016
📖 §9.4 Key Elements of M&T

18. The essential elements of monitoring and targeting system is

  1. Recording
  2. Reporting
  3. Controlling
  4. All of the above
Answer: D) All of the above
Confirmed vs Book-1 §9.4 — the key elements are Recording, Analysing & Comparing, Setting Targets, Monitoring, Reporting and Controlling. Recording, Reporting and Controlling are all listed, so 'all of the above'. Answer (d).
Source: 2016
📖 §9.6 Linear Regression — E = C + M·P

19. In an industry the billed electricity consumption for a month is 5.8 lakh kWh. The fixed electricity consumption of the plant is 30000kWh and with a variable electricity consumption of 11 kWh/ton. Calculate the production of the industry

  1. 50000 tonnes
  2. 60000 tonnes
  3. 58000 tonnes
  4. None of the above
Answer: A) 50000 tonnes
Confirmed vs Book-1 §9.6 — E = C + M·P, so P = (E - C)/M = (5,80,000 - 30,000)/11 = 5,50,000/11 = 50,000 tonnes. The fixed 30,000 kWh must be deducted before dividing by the variable rate. Answer (a).
Source: 2016
📖 §9.6 CUSUM Charts (Fig 9.12)

20. In a cumulative sum chart if the graph is going up, it means

  1. Energy consumption is going up
  2. Energy consumption is going down
  3. Specific energy consumption is coming down
  4. No inference can be made
Answer: A) Energy consumption is going up
Confirmed vs Book-1 §9.6 — CUSUM = Σ(E_act - E_calc); a RISING line means actual consistently exceeds the calculated/target consumption, i.e. energy consumption (and specific energy consumption) is going up - performance is worsening due to poor control, housekeeping or maintenance. A falling line indicates savings. Answer (a).
Source: 2016
📖 Project scheduling — Gantt chart (outside Ch-9 §9.1-9.7)

21. The technique used for scheduling the tasks and tracking of the progress of energy management projects through a bar chart is called

  1. CPM
  2. Gantt chart
  3. CUSUM
  4. PERT
Answer: B) Gantt chart
Confirmed — Book-1 (project monitoring, outside the Ch-9 text): a Gantt chart is the bar-chart technique for scheduling tasks and tracking progress of energy-management projects against time. CPM/PERT are network techniques and CUSUM is a cumulative-deviation energy chart, not a scheduling tool. Answer (b).
Source: 2013
📖 §9.6 CUSUM Charts (Book EOC Objective Q3)

22. In a cumulative sum (CUSUM) chart, if the graph is going up, then

  1. nothing can be said
  2. actual and calculated energy consumption are the same
  3. energy consumption is reduced
  4. specific energy consumption is going up
Answer: D) specific energy consumption is going up
Confirmed vs Book-1 §9.6 — CUSUM = Σ(E_act - E_calc). A rising CUSUM line means actual energy exceeds the calculated (production-normalised) energy month after month, so the specific energy consumption is going up, i.e. performance is deteriorating. A horizontal line means actual = calculated. Answer (d).
Source: 2013
📖 §9.6 Linear Regression — E = C + M·P

23. The fixed energy consumption of a company is 2000 kWh per month. The line slope of the energy (y) versus production (x) chart is 0.3. The energy consumed in kWh per month for a production level of 80,000 tons/month is

  1. 24,000 kWh
  2. 24,200 kWh
  3. 26,000 kWh
  4. 38,000 kWh
Answer: C) 26,000 kWh
Confirmed vs Book-1 §9.6 — E = 0.3 × 80,000 + 2,000 = 24,000 + 2,000 = 26,000 kWh/month. Option (a) 24,000 is the variable part only and omits the fixed base load C. Answer (c).
Source: 2013
📖 §9.6 y = c + mx (E = M·P + C)

24. The empirical relationship used to plot Production Vs Energy consumption is……………… ( where Y= energy consumed for the period; C = fixed energy consumption; M = energy consumption directly related to production; X= production for same period).

  1. X=Y+MC
  2. Y=MX+C
  3. M=CX+Y
  4. Y= MX-C
Answer: B) Y=MX+C
Confirmed vs Book-1 §9.6 — the best-fit line is y = c + mx, i.e. Energy for the period = C + M × Production for the same period, written Y = MX + C. C (intercept) = fixed/base-load energy, M (slope) = production-related (variable/specific) energy. Answer (b).
Source: 2012
📖 Financial management — depreciation & cash flow (outside Ch-9 §9.1-9.7)

25. If the asset depreciation is considered, then net operating cash inflow would be

  1. higher
  2. lower
  3. no effect
  4. none of these
Answer: B) lower
Confirmed as printed — this item is from the financial-management syllabus, not Book-1 Ch-9; nothing in §9.1-9.7 contradicts or supports it. Depreciation is a non-cash charge that lowers the reported net operating profit/inflow figure once it is deducted (the marked key is 'lower'), even though in post-tax cash-flow analysis it is added back as a tax shield. Answer as keyed (b).
Source: 2012
📖 §9.6 XY Scatter Diagram (Book EOC Objective Q1)

26. A chart in Scatter Diagram shows a low degree of scatter. It is indicative of

  1. good fit
  2. poor fit
  3. skewed fit
  4. normal fit
Answer: A) good fit
Confirmed vs Book-1 §9.6 — 'This chart shows a low degree of scatter indicative of a good fit.' Low scatter = points lie close to the best-fit line = high correlation coefficient (r near 1) = good control of energy use. Answer (a).
Source: Guidebook
📖 §9.6 Linear Regression Analysis (Book EOC Objective Q2)

27. ______ is a statistical technique which determines and quantifies the relationship between variables and enables standard equations to be established for energy consumption.

  1. linear regression analysis
  2. time-dependent energy analysis
  3. moving annual total
  4. CUSUM
Answer: A) linear regression analysis
Confirmed vs Book-1 §9.6 — 'Linear regression analysis is a statistical technique which determines and quantifies the relationship between variables and enables standard equations to be established for energy consumption', giving the standard performance line E = M·P + C. Answer (a).
Source: Guidebook
📖 §9.1 Principle of M&T (Book EOC Objective Q4)

28. Energy monitoring and targeting is built on the principle of

  1. production can be reduced to achieve reduced energy consumption
  2. Consumption of energy is proportional to production rate
  3. You cannot manage what you do not measure
  4. None of the above
Answer: C) You cannot manage what you do not measure
Confirmed vs Book-1 §9.1 — M&T 'is based on the principle "you can't manage what you don't measure"'. It combines the principles of energy use and statistics and manages utilities as controllable resources. Answer (c).
Source: Guidebook
📖 §9.6 Factors influencing energy consumption (Book EOC Objective Q5)

29. Which of the variable does not contribute to energy consumption?

  1. production
  2. hours
  3. climate
  4. none of the above
Answer: D) none of the above
Confirmed vs Book-1 §9.6 / Table 9.4 — production volume, operating hours and climate all influence energy consumption and are legitimate normalising variables. Since every listed variable does contribute, the correct choice is 'none of the above'. Answer (d).
Source: Guidebook
📖 §9.6 XY Scatter Diagram (Book EOC Objective Q6)

30. Poor scattering on trend line of production Vs Energy consumption indicates

  1. poor level of control
  2. good level of control
  3. both the above
  4. none of above
Answer: A) poor level of control
Corrected (was b) — Book-1 §9.6: 'If data fit is poor, it indicates poor level of control and hence a scope for energy savings... if it is known that there should be a relationship, it indicates a poor level of control and hence a potential for energy savings.' Poor scattering about the production-vs-energy trend line therefore means a POOR level of control (high savings potential); low scatter would mean good control. Answer (a).
Source: Guidebook
📖 §9.6 XY Scatter / y-intercept (Book EOC Objective Q7)

31. Fixed energy consumption can be determined from a

  1. bar chart
  2. vertical line chart
  3. pie chart
  4. XY coordinate system
Answer: D) XY coordinate system
Confirmed vs Book-1 §9.6 — fixed (base-load) energy is the intercept C of the best-fit line E = M·P + C, read where the line cuts the y-axis on an XY (scatter) coordinate plot. Bar charts and pie charts show shares and monthly totals only and cannot yield an intercept. Answer (d).
Source: Guidebook
📖 §9.6 Graphical correlation of energy & production (Book EOC Objective Q8)

32. The best way of correlating production and energy data in any plant is

  1. text format
  2. graphical representation
  3. oral communication
  4. none
Answer: B) graphical representation
Confirmed vs Book-1 §9.6 — the guidebook correlates production and energy graphically (XY scatter with best-fit line), which reveals the slope M (variable energy), the intercept C (fixed energy) and the degree of scatter (level of control). Text or oral formats reveal none of these. Answer (b).
Source: Guidebook
📖 §9.6 Relating energy consumption to production (Book EOC Objective Q9)

33. For any company, energy consumption mostly relates to

  1. profits
  2. inventory
  3. production
  4. all the above
Answer: C) production
Confirmed vs Book-1 §9.4/§9.6 — energy consumption is related to a measured output, normally production quantity; the whole M&T method rests on the energy-production relationship E = M·P + C. Profits and inventory are not energy drivers. Answer (c).
Source: Guidebook
📖 §9.6 CUSUM Charts (Book EOC Objective Q10)

34. In a cumulative sum chart, if the graph is horizontal, then

  1. nothing can be said
  2. energy consumption is reduced
  3. specific energy consumption is increasing
  4. actual and calculated energy consumption are the same
Answer: D) actual and calculated energy consumption are the same
Confirmed vs Book-1 §9.6 — CUSUM = Σ(E_act - E_calc). A horizontal line means the running sum is not changing, i.e. each month's difference is zero, so actual and calculated (target) energy consumption are the same and performance is on target. Answer (d).
Source: Guidebook
📖 §9.6 Linear Regression — E = C + M·P

35. In an industry the electricity consumed for a period is 1,10,000 kWh. The production in the period is 12,000 tons with a variable energy consumption of 6 kWh/Ton. The fixed kWh of the plant is ____.

  1. 35000
  2. 38000
  3. 32000
  4. 36000
Answer: B) 38000
Confirmed vs Book-1 §9.6 — variable energy = 6 kWh/ton × 12,000 tons = 72,000 kWh. Fixed C = 1,10,000 - 72,000 = 38,000 kWh, the base load independent of output. Answer (b).
Source: Mar 2023
📖 M&V savings formula (outside Ch-9 §9.1-9.7)

36. Formula for computing energy savings as part of Measurement & Verification is ____.

  1. Energy Savings = Base year energy use + post-retrofit energy use +/- Adjustments
  2. Energy Savings = Base year energy use - post-retrofit energy use +/- Adjustments
  3. Energy Savings = post-retrofit energy use - base year energy use +/- Adjustments
  4. None of the above
Answer: B) Energy Savings = Base year energy use - post-retrofit energy use +/- Adjustments
Confirmed — Measurement & Verification convention (outside the Ch-9 text but consistent with §9.4 baseline practice): Energy Savings = Baseline (base-year) energy use - Post-retrofit energy use ± Adjustments, the adjustments normalising for production, weather and other changed conditions. Answer (b).
Source: Mar 2023
📖 §9.6 Normalising factors / Table 9.4

37. Which of the following is not a common normalizing factor for industrial facilities?

  1. Input
  2. Output
  3. Product type
  4. Maintenance cost
Answer: D) Maintenance cost
Confirmed vs Book-1 §9.6 and Table 9.4 — genuine normalising factors are those that physically drive energy use: output (production volume), input (raw material/steam/air delivered) and product type/mix. Maintenance cost is a financial figure, not an energy driver, so it is NOT a normalising factor. Answer (d).
Source: Mar 2023
📖 §9.6 Table 9.4 Factors influencing energy consumption

38. Factors influencing energy consumption in an organization include ____.

  1. Operational Hours
  2. Units of Production
  3. Usage Behavior
  4. All of the above
Answer: D) All of the above
Confirmed vs Book-1 §9.6/Table 9.4 — operational hours, units of production and usage behaviour (operating practice/housekeeping) all influence energy consumption; regression in M&T is built on such influencing variables. Answer (d).
Source: Mar 2023
📖 §9.6 Specific Energy Consumption (Figs 9.7-9.8)

39. Specific energy consumption is defined as:

  1. Energy consumption per month
  2. Annual energy consumption
  3. Energy consumed per unit of fuel burnt
  4. Energy consumed per unit of production
Answer: D) Energy consumed per unit of production
Confirmed vs Book-1 §9.6 — SEC is the energy consumed per unit of production (e.g. kWh/tonne, toe/tonne); it is plotted monthly to reveal trends. Energy per month or per year is simply consumption, not a specific figure. Answer (d).
Source: Jul 2022
📖 §9.6 XY Scatter / y-intercept

40. Fixed energy consumption can be determined from:

  1. Bar chart
  2. Vertical line chart
  3. Pie chart
  4. XY coordinate system
Answer: D) XY coordinate system
Confirmed vs Book-1 §9.6 — plotting energy (y) against production (x) on an XY coordinate system and extending the best-fit line to zero production gives the intercept C = fixed energy consumption. Answer (d).
Source: Sep 2025
📖 §9.4 Standard energy performance / baseline

41. Why is an energy baseline established in Monitoring and Targeting (M&T)?

  1. To record only monthly electricity bills
  2. To fix a reference point for measuring energy performance improvements
  3. To eliminate the need for energy performance indicators
  4. To avoid sharing information with managers and stakeholders
Answer: B) To fix a reference point for measuring energy performance improvements
Confirmed vs Book-1 §9.4 — 12-24 months of energy and output data are regressed to obtain the standard energy performance, which 'provides a base line for the assessment of future performance' and can be used as an initial target. The baseline is therefore the reference for measuring improvement. Answer (b).
Source: Sep 2025
📖 §9.6 CUSUM Charts

42. In a cumulative sum chart, a horizontal graph indicates:

  1. Nothing can be said
  2. Energy consumption is reduced
  3. Specific energy consumption is increasing
  4. Actual and calculated energy consumption are the same
Answer: D) Actual and calculated energy consumption are the same
Confirmed vs Book-1 §9.6 — a flat (horizontal) CUSUM means the running Σ(E_act - E_calc) is unchanged, i.e. actual equals calculated consumption each period and performance is exactly on target. Rising = worsening, falling = savings. Answer (d).
Source: Sep 2025
📖 §9.1 Principle of M&T

43. Which principle is energy monitoring and targeting based on?

  1. Energy consumption is constant
  2. You can't manage what you don't measure
  3. Energy consumption is unpredictable
  4. Production rate has no effect
Answer: B) You can't manage what you don't measure
Confirmed vs Book-1 §9.1 — M&T 'is based on the principle "you can't manage what you don't measure"', combining energy-use principles with statistics. Answer (b).
Source: Sep 2024
📖 §9.6 Specific Energy Consumption (Table 9.1 basis)

44. A manufacturing plant consumes 5 tonnes of coal (CV = 4000 kCal/kg) to produce 25 tonnes of cement. The Specific Energy Consumption (SEC) of the plant shall be

  1. 100 kcal/kg of cement
  2. 200 kcal/kg of cement
  3. 400 kcal/kg of cement
  4. 800 kcal/kg of cement
Answer: D) 800 kcal/kg of cement
Confirmed vs Book-1 §9.6 — energy input = 5 t × 1000 kg/t × 4000 kCal/kg = 2 × 10^7 kCal. SEC = energy / output = 2 × 10^7 kCal / 25,000 kg of cement = 800 kCal/kg of cement. Answer (d).
Source: Sep 2024
📖 Book-1 §9.1 Introduction; Ch-9 Objective Q4

45. Monitoring and Targeting (M&T) is fundamentally based on which management principle?

  1. Energy saved is energy generated
  2. You can't manage what you don't measure
  3. Prevention is better than cure
  4. Reduce, reuse, recycle
Answer: B) You can't manage what you don't measure
Confirmed vs Book-1 §9.1 — M&T "is based on the principle 'you can't manage what you don't measure'", which is guidebook Ch-9 Objective Q4. M&T uses energy information to eliminate waste, reduce and control current energy use and improve operating procedures, combining energy use with statistics. The book records typical reductions in annual energy costs of 5–15%.
Source: AI practice
📖 Book-1 §9.6 Linear Regression Analysis — XY scatter, y = c + mx; Ch-9 Objective Q7

46. In the energy-production relationship E = m·P + c, what does the intercept 'c' represent?

  1. Variable (specific) energy consumption
  2. The slope of the best-fit line
  3. Fixed energy consumption (base load)
  4. The correlation coefficient
Answer: C) Fixed energy consumption (base load)
Confirmed vs Book-1 §9.6 — "y = c + mx ... c is the value at which the straight-line curve intersects the y axis, and m is the gradient", i.e. Energy consumed = C + m × production. In the book’s foundry example E = 0.4P + 180, "the theoretical base load for furnace is 180 mtoe" — so the intercept is the FIXED energy consumption. Guidebook Objective Q7 confirms fixed energy consumption is read off an XY coordinate system.
Source: AI practice
📖 Book-1 §9.6 Linear Regression Analysis — Example 9.1 (foundry, E = 0.4P + 180)

47. A foundry's energy-production data fits the line E = 0.4P + 180 (E in toe, P in tonnes). What is the variable (specific) energy consumption?

  1. 180 toe
  2. 0.4 toe per tonne
  3. 72 toe
  4. 220 toe per tonne
Answer: B) 0.4 toe per tonne
Confirmed vs Book-1 §9.6 — this is the book’s own foundry regression result, derived from the normal equations as m = 0.4 and c = 180. The gradient m = 0.4 toe per tonne is the variable (specific) energy consumption — the extra energy needed per extra tonne produced. The intercept 180 toe is the fixed / base-load energy, consumed even at zero production, so (a) and (d) confuse the two terms.
Source: AI practice
📖 Book-1 §9.6 CUSUM Charts (Figures 9.12 / 9.13); Ch-9 Objectives Q3 and Q10

48. On a CUSUM chart, a line that trends steadily DOWNWARD indicates that the plant is:

  1. Consuming more energy than predicted (worsening)
  2. Achieving energy savings (consuming less than predicted)
  3. Exactly on target
  4. Producing at maximum capacity
Answer: B) Achieving energy savings (consuming less than predicted)
Confirmed vs Book-1 §9.6 — CUSUM is the cumulative sum of (actual − calculated) energy, the calculated value coming from the baseline equation E = mP + c. The book reads a rising line as declining performance (Objective Q3: graph going up → specific energy consumption going up) and a horizontal line as actual = calculated (Objective Q10). In the 18-month foundry example the line drops after month 11 and the accumulated fall of 44 toe (50 − 6) is reported as the saving — so a falling line means savings.
Source: AI practice
📖 Book-1 §9.6 CUSUM — chapter Solved Example (E_calc = 0.5P + 220, July 2011 row)

49. A baseline equation is E_calc = 0.5P + 220 (toe). In a given month actual energy E_act = 590 toe at production P = 760 tonnes. The CUSUM contribution (diff) for this month is:

  1. +10 toe
  2. -10 toe
  3. +600 toe
  4. -370 toe
Answer: B) -10 toe
Confirmed vs Book-1 §9.6 — these are the July figures of the guidebook’s waste-heat-recovery solved example, where the printed difference is −10 toe. Working: E_calc = 0.5 × 760 + 220 = 380 + 220 = 600 toe; difference = E_act − E_calc = 590 − 600 = −10 toe. A negative difference means less energy was used than predicted (a saving) and it is carried into the running CUSUM total, which reaches −96 toe by December.
Source: AI practice
📖 Book-1 §9.6 XY Scatter Diagram / Correlation Coefficients; Ch-9 Objectives Q1 and Q6

50. In an XY scatter plot of energy versus production, poor (wide) scatter of the data points around the best-fit line indicates:

  1. Excellent process control
  2. Poor control and significant scope for energy savings
  3. A perfect correlation coefficient of 1
  4. That production has no effect on energy
Answer: B) Poor control and significant scope for energy savings
Confirmed vs Book-1 §9.6 — "If data fit is poor, it indicates poor level of control and hence a scope for energy savings", which is guidebook Objective Q6 (poor scattering → poor level of control). Conversely a low degree of scatter is "indicative of a good fit" (Objective Q1) and of consistent control. The Pearson correlation coefficient r lies between 0 and 1, with 1 representing 100% correlation; the book’s foundry example gives r = 0.98.
Source: AI practice
📖 Book-1 §9.6 Linear Regression Analysis (Table 9.4) and §9.7 EMIS features; Ch-9 Objective Q5

51. Which of the following is NOT a valid normalizing factor for specific energy consumption?

  1. Production output
  2. Degree-days (weather)
  3. Maintenance cost
  4. Hours of occupancy
Answer: C) Maintenance cost
Confirmed (option (d) repaired) — Book-1 §9.6/§9.7: the variables the guidebook actually regresses energy against are units of production, hours of occupancy (for lighting) and, in EMIS, "other related variables such as degree days and production data"; Table 9.4 lists air volume delivered, steam generated and production volume. Maintenance cost appears nowhere as an influencing variable — it is a cost, not a driver of energy use — so it is the odd one out. Option (d) previously read "Product type", which the 2014 Chapter 9 does not list either and so risked a second correct answer; it is replaced by the book’s own "hours of occupancy". Answer unchanged at (c).
Source: AI practice
📖 Book-1 §4.7 Energy Performance — Production Factor and Reference Year Equivalent (applied to Ch-9 M&T)

52. A plant used 5000 toe to make 100,000 t in the reference year. In the current year it made 90,000 t. What is the Production Factor and the reference-year-equivalent energy?

  1. PF = 1.11; 5555 toe
  2. PF = 0.90; 4500 toe
  3. PF = 0.90; 5000 toe
  4. PF = 0.10; 500 toe
Answer: B) PF = 0.90; 4500 toe
Confirmed vs Book-1 §4.7 — "Production factor is the ratio of production in the current year to that in the reference year" and "Reference year equivalent = Reference year energy use × Production factor". Working: PF = 90,000 / 1,00,000 = 0.90; reference-year equivalent = 5,000 × 0.90 = 4,500 toe. Note the ref: the guidebook develops production factor and plant energy performance in Chapter 4 §4.7, and Chapter 9 M&T applies the same normalisation idea.
Source: AI practice
📖 Book-1 §4.7 Energy Performance — Plant Energy Performance (applied to Ch-9 M&T)

53. Using a reference-equivalent energy of 4500 toe, if the plant's actual current-year energy is 4200 toe, the Plant Energy Performance is:

  1. -6.7% (worsened)
  2. +6.7% (improved)
  3. 0% (no change)
  4. +93.3%
Answer: B) +6.7% (improved)
Confirmed vs Book-1 §4.7 — Plant energy performance = (Reference year equivalent − Current year’s energy) / Reference year equivalent × 100. Working: (4,500 − 4,200)/4,500 × 100 = 300/4,500 × 100 = +6.7%. The book states "the greater the improvement, the higher the number will be", so a POSITIVE value is an improvement and a negative value a deterioration from the reference year.
Source: AI practice
📖 Book-1 §9.4 Key Elements of Monitoring & Targeting System

54. The correct order of the six key elements of a Monitoring and Targeting system is:

  1. Targeting → Recording → Reporting → Monitoring → Analysing → Controlling
  2. Recording → Analysing & Comparing → Setting Targets → Monitoring → Reporting → Controlling
  3. Monitoring → Recording → Controlling → Reporting → Targeting → Analysing
  4. Reporting → Recording → Monitoring → Targeting → Analysing → Controlling
Answer: B) Recording → Analysing & Comparing → Setting Targets → Monitoring → Reporting → Controlling
Confirmed vs Book-1 §9.4 — the six key elements are listed in this order: Recording, Analysing & Comparing, Setting Targets, Monitoring, Reporting, Controlling. Analysing & Comparing uses 12–24 months of historical data and regression to fix the standard energy performance, which then serves as the baseline and initial target. Energy Account Centres (§9.3) are established first so managers are accountable, and §9.7 recommends "reporting by exception" so managers are not swamped.
Source: AI practice
📖 Book-1 §8.8 Measurement and Verification (also §7.8 Calculating savings)

55. In Measurement & Verification (M&V), energy savings from a retrofit are best expressed as:

  1. Post-retrofit energy minus base-year energy
  2. Base-year (pre-retrofit) energy minus post-retrofit energy, plus or minus adjustments
  3. Base-year energy divided by post-retrofit energy
  4. Production multiplied by specific energy consumption
Answer: B) Base-year (pre-retrofit) energy minus post-retrofit energy, plus or minus adjustments
Confirmed vs Book-1 §8.8 — "Energy Savings = Base year Energy Use − Post-Retrofit Energy Use ± Adjustments"; §7.8 states the same as Energy Saved = Baseline − Current ± Adjustment. The adjustments term "brings energy use in the two time periods to the same set of conditions" — commonly weather, occupancy, plant throughput and required equipment operation — and may be positive or negative. Option (a) reverses the subtraction and would report a saving as a negative number.
Source: AI practice
📖 Book-1 §9.6 Data and Information Analysis — Moving Annual Total (Figure 9.9)

56. Which graphical M&T tool plots each point as the sum of the previous 12 months' consumption to smooth out seasonal variation and meter-timing errors?

  1. Time-dependent graph
  2. Deviance chart
  3. Moving Annual Total (MAT)
  4. Norm chart
Answer: C) Moving Annual Total (MAT)
Confirmed vs Book-1 §9.6 — "For this chart, each point represents the sum of previous 12 months of data. In this way, each point has full range of the seasons, holidays. The technique also smoothes out errors in the timings of meter readings." The distractors are the book’s other tools: the time-dependent graph shows trends but cannot explain them, the norm chart overlays actual on target consumption, and the deviance chart plots the difference between target and actual. When MAT energy and production track each other, "it suggests that there is no cause for alarm".
Source: AI practice

Short questions (5 marks) — 44

📖 §9.1-9.2 Definition of Energy Monitoring & Targeting

1. What do you understand by Energy Monitoring and Targeting (M&T)?

Model answer: Energy Monitoring and Targeting (M&T) is primarily a management technique that uses energy information as the basis to eliminate waste, reduce and control the current level of energy use, and improve existing operating procedures. It combines the principles of energy use and statistics and is based on the principle 'you can't manage what you don't measure'. Monitoring establishes the existing pattern of consumption and explains deviations from it; targeting identifies a desirable consumption level and works towards achieving it. M&T typically reduces annual energy costs by 5-15%.
Verbatim source: Sec 9.1 'primarily a management technique that uses energy information as a basis to eliminate waste...' and the 5-15% figure.
Source: BEE Guidebook 2014
📖 §9.6 Correlation Coefficients (Pearson r, Table 9.5)

2. What is meant by correlation coefficient? How is it useful in energy monitoring?

Model answer: The Pearson correlation coefficient (r) indicates how well the best-fit straight line correlates to the scattered sample data, i.e. the reliability of the line drawn. It is a value between 0 and 1, with 1 representing 100% correlation (the book's worked foundry example gives r = 0.98, which is very good). In energy monitoring it confirms whether the chosen driver (e.g. production) reliably explains energy consumption, validating the regression-based standard equation used for targeting. A high r (low scatter) means good control; a low r (poor scatter) means poor control and hence scope for energy savings. The minimum acceptable r decreases as the number of data points increases (10 points need 0.767, 30 points need 0.464).
Source states r is 'a value between 1 and 0, with a value of 1 representing 100% correlation' and gives r = 0.98; Table 9.5 lists minimum r values.
Source: BEE Guidebook 2014
📖 §9.1 Need for a monitoring programme

3. What is the need for a monitoring system?

Model answer: An energy audit only produces a 'picture' of past energy consumption; to keep control of subsequent consumption a monitoring programme is needed. A monitoring system regularly measures and records the actual energy consumption of each Energy Account Centre, relates it to a measured output such as production, and compares actual use against expected values/targets so deviations are detected promptly. It checks accuracy of energy invoices, allocates energy costs to departments, highlights performance problems in equipment/systems, and lets energy-efficiency projects be verified for results. In short, you cannot manage and improve what you do not measure.
Combines Sec 9.1 (audit gives only a past picture, monitoring needed to keep control) with the 'M&T system will involve' list in Sec 9.4.
Source: BEE Guidebook 2014
📖 §9.6 CUSUM — Steps for CUSUM analysis

4. List at least 5 steps involved in CUSUM analysis.

Model answer: (1) Plot the Energy-Production graph for the baseline (pre-intervention) months and draw the best-fit straight line. (2) Derive the equation of the line, E = mP + c (e.g. the book's E = 0.4P + 180). (3) Calculate the standard/calculated energy consumption (E_calc) for each month from the equation using actual production. (4) Calculate the difference between actual and standard consumption, E_act - E_calc, for each period. (5) Compute the CUSUM as the running cumulative sum of these differences. (6) Plot the CUSUM graph against time and estimate the savings accumulated from the energy-saving measure.
Source 'Steps for CUSUM analysis' list 1-8 (foundry heat-recovery example, E = 0.4P + 180).
Source: BEE Guidebook 2014
📖 §9.6 Single-variable vs multi-variable regression

5. Explain the difference between single variable and multiple variable analysis.

Model answer: In single-variable analysis, energy consumption (y) is related to only one independent variable (x), typically production, giving the simple linear relationship y = c + mx; it is suitable where one factor dominates energy use and can be solved by hand using the normal equations. In multi-variable analysis, energy is influenced by several different variables simultaneously, described by y = c + m1x1 + m2x2 + ... + mn xn (e.g. production, degree-days, occupancy hours). Multivariable analysis is difficult to solve by hand calculation, so specialist computer software is advised to determine the statistical relationship between the variables.
Source: y = c + mx for single variable; multi-variable 'y = ... + mn xn' with note that it is difficult by hand and needs specialist computer software.
Source: BEE Guidebook 2014
📖 §9.1 Core principle of M&T

6. State the core principle on which energy monitoring and targeting is based, and what it essentially combines.

Model answer: M&T is based on the principle 'you can't manage what you don't measure'. It is essentially a management technique that combines the principles of energy use and statistics. By using M&T, all plant and building utilities (fuel, steam, refrigeration, compressed air, water, effluent, electricity) are managed as controllable resources in the same way as raw materials, inventory, occupancy, personnel and capital. M&T programmes have shown typical reductions in annual energy costs of between 5 and 15% across industrial sectors.
Direct from Sec 9.1; objective Q4 confirms the principle wording.
Source: BEE Guidebook 2014
📖 §9.2 Monitoring vs Targeting

7. Differentiate between 'Monitoring' and 'Targeting' in an M&T programme.

Model answer: Monitoring is the process of establishing the existing pattern of energy consumption and explaining deviations from that pattern; its primary goal is to maintain the existing pattern by providing all the necessary data on energy consumption and key related data such as production. Targeting is the identification of a desirable energy consumption level and working towards achieving it. Targets are based on the historical (average or best) data acquired during monitoring, as well as benchmarking with the energy performance of similar organisations.
Verbatim definitions from Sec 9.2.
Source: BEE Guidebook 2014
📖 §9.3 Energy Account Centres (EACs)

8. What is an Energy Account Centre (EAC) and why is it established before initiating M&T?

Model answer: Before initiating M&T it is important to establish Energy Account Centres (EACs) within an organisation. EACs may be departments, processes or cost centres. Operational managers should be made accountable for the energy consumption of the EACs for which they are responsible. EACs are defined according to the site/metering arrangement: a single site with central metering is best treated as a single EAC, while sub-metering allows a site to be broken up into several separate EACs; multiple sites with central meters are each treated as separate EACs.
Source Sec 9.3 'establish Energy Account Centers (EACs)... departments, processes or cost centers' and the site/metering classification.
Source: BEE Guidebook 2014
📖 §9.4 Key elements of an M&T system

9. List and briefly describe the key elements of a Monitoring & Targeting system.

Model answer: (1) Recording - measuring and recording energy consumption of each EAC by setting up procedures for regular collection of reliable data. (2) Analysing & Comparing - relating energy consumption to a measured output (e.g. production) over 12-24 months to obtain standard energy performance via regression. (3) Setting Targets - setting achievable targets that improve on standard energy performance. (4) Monitoring - comparing actual consumption to the set target on a regular basis. (5) Reporting - reporting results and variances to management. (6) Controlling - implementing management measures to correct any variances.
Source Sec 9.4 lists exactly these six elements in this order.
Source: BEE Guidebook 2014
📖 §9.4 Standard energy performance (12-24 months, regression)

10. What is 'standard energy performance' and how is it established?

Model answer: Standard energy performance is obtained by relating energy consumption to a measured output, such as production quantity, using 12-24 months of historical data for each EAC. It is established through regression analysis of past data; if such data do not exist, an energy audit is conducted to establish it. Standard energy performance provides a baseline for the assessment of future performance and can also be used as an initial target. Energy cost savings are consistently achieved when improvements are made on the standard energy performance.
Source Sec 9.4 'Analysing & Comparing' element specifies 12-24 months, regression, baseline and initial target.
Source: BEE Guidebook 2014
📖 §9.4 Benefits of M&T

11. List the benefits of a Monitoring & Targeting programme.

Model answer: The ultimate goal is to reduce energy costs through improved energy efficiency and management control. Other benefits: identify and explain an increase or decrease in energy use; draw energy consumption trends (weekly, seasonal, operational); improve energy budgeting in line with production plans; observe how the organisation reacted to past changes; determine future energy use when planning operational changes; diagnose specific areas of wasted energy; develop performance targets for energy management programmes; and manage energy consumption rather than accept it as a fixed, uncontrollable cost.
Source Sec 9.4 'Benefits of M&T' bullet list.
Source: BEE Guidebook 2014
📖 §9.4 Activities involved in an M&T system

12. What activities does an M&T system particularly involve?

Model answer: An M&T system particularly involves: checking the accuracy of energy invoices; allocating energy costs to specific departments (Energy Accounting Centres); determining energy performance/efficiency; recording energy use so that projects intended to improve energy efficiency can be checked for results; and highlighting performance problems in equipment or systems.
Source Sec 9.4 'M&T system will involve the following' bullet list.
Source: BEE Guidebook 2014
📖 §9.5 Data and information sources

13. From what sources can information related to energy use be obtained in an organisation?

Model answer: Plant-level information can be derived from financial accounting systems (the utilities cost centre). Plant/department-level information can be found in comparative energy consumption data for a group of similar facilities and service-entrance meter readings. System-level performance data (e.g. compressor house) is determined from sub-metering data. Equipment-level information is obtained from nameplate data, run-time and schedule information, and sub-metered data on specific energy-consuming equipment. All such data can be processed to yield information about facility performance.
Source Sec 9.5 lists plant, plant-department, system and equipment level sources.
Source: BEE Guidebook 2014
📖 §9.6 Annual energy consumption analysis (Table 9.1-9.3, Figs 9.1-9.2)

14. Describe the simplest data-analysis technique for assessing annual energy consumption, and state its limitation.

Model answer: The simplest analysis is to produce a percentage breakdown of annual energy consumption and cost data. The steps are: convert all energy data into standard units (usually kcal) using standard conversion factors (e.g. electricity 860 kcal/kWh, HSD 10,500, furnace oil 10,200, LPG 12,000 kcal/kg); compile annual consumption and cost for each fuel; produce a percentage breakdown; and draw pie charts of energy and cost share. The limitation is that it makes no allowance for variable factors (climatic zone, building occupancy), so it cannot be used as a comparison tool between different organisations.
Source Sec 9.6 + Table 9.1 conversion factors; explicit limitation that it 'cannot be used as a comparison tool between different organizations'.
Source: BEE Guidebook 2014
📖 §9.6 Table 9.1 Standard Energy Conversions

15. State the standard energy conversion factors (to kcal) used in M&T annual energy analysis.

Model answer: Electricity: 1 kWh = 860 kcal. HSD (High Speed Diesel): 1 kg = 10,500 kcal. Furnace Oil: 1 kg = 10,200 kcal. LPG: 1 kg = 12,000 kcal. All energy consumption data are converted to these standard units (kcal) so that different fuels and energy types can be summed and compared on a common basis.
Source Table 9.1 'Standard Energy Conversions'.
Source: BEE Guidebook 2014
📖 §9.6 Time-dependent energy analysis (Fig 9.4)

16. What is time-dependent energy analysis and what are its limitations?

Model answer: If monthly energy consumption data are collected, a simple graph of energy consumption plotted against time can be produced. This time-dependent analysis identifies general trends and seasonal patterns and lets exceptions to the norm be spotted immediately; more than one variable (e.g. oil with electricity) can be plotted together. Its limitation is that it is difficult to find out why certain trends occur or whether a particular trend really exists; it can only be used as a comparative tool, not an absolute one, and further analysis (e.g. regression) is needed to explain the trends.
Source Sec 9.6 time-dependent analysis paragraph including 'comparative tool and not an absolute one'.
Source: BEE Guidebook 2014
📖 §9.6 Norm chart (Fig 9.5)

17. What is a norm chart and what is its main use and limitation?

Model answer: A norm chart is a sequential plot of actual energy consumption overlaid on a plot of target (normal) consumption. It is of little value as an analytical tool, but is useful for highlighting exceptions and communicating them to managers. Because norm charts represent a historical record of energy consumption, senior and operational managers find them relatively easy to understand.
Source Sec 9.6 'Norm Chart' paragraph (Figure 9.5).
Source: BEE Guidebook 2014
📖 §9.6 Deviance chart (Fig 9.6)

18. What is a deviance chart and how is it interpreted?

Model answer: A deviance chart plots the difference between target and actual energy consumption. If, in a given month, consumption is above the target it is plotted as a positive value; if actual consumption is below the predicted value, a negative value is returned. It is useful to show the limits of normal operation on the graph to distinguish normal variation from serious deviations. Deviance charts are particularly good at highlighting problems so that remedial action can be taken, and can be used to initiate detailed exception reports.
Source Sec 9.6 'Deviance Chart' (Figure 9.6).
Source: BEE Guidebook 2014
📖 §9.6 Moving Annual Total (Fig 9.9)

19. What is a Moving Annual Total (MAT) and what are its advantages?

Model answer: A Moving Annual Total represents energy and production data such that each plotted point equals the sum of the previous 12 months of data; 12 months of energy and production data are needed to start the chart. Because each point covers a full range of seasons and holidays, the technique removes seasonal effects and also smooths out errors in the timing of meter readings. If energy and production lines track each other there is no cause for alarm; any deviation in the energy line gives early warning of energy waste or confirms that efficiency measures are having a positive impact.
Source Sec 9.6 'Moving Annual Total' (Figure 9.9).
Source: BEE Guidebook 2014
📖 §9.6 Linear regression analysis

20. What is linear regression analysis in energy management, and how does it overcome the limitation of time-dependent analysis?

Model answer: Linear regression analysis is a statistical technique that determines and quantifies the relationship between variables, enabling standard equations for energy consumption to be established from data that would otherwise be meaningless. It overcomes the limitation of time-dependent analysis by removing the 'time' element and focusing instead on the variables that influence energy consumption (e.g. furnace-oil or electricity versus units of production, lighting energy versus occupancy hours). Its reliability depends heavily on the quantity and quality of data used, so results should be treated with care.
Source Sec 9.6 'Linear Regression Analysis' paragraph.
Source: BEE Guidebook 2014
📖 §9.6 XY Scatter Diagram (Fig 9.10)

21. What does an XY scatter diagram of energy versus production reveal, and how is the degree of scatter interpreted?

Model answer: An XY scatter diagram (energy as the dependent y-variable, production as the independent x-variable) gives more understanding of the relationship between energy and production and yields a best-fit straight line E = c + mP. A low degree of scatter indicates a good fit and a good level of control. If the data fit is poor while a relationship is expected, it indicates a poor level of control and hence scope/potential for energy savings. The fixed energy consumption (base load) is read as the intercept where the best-fit line cuts the y-axis on the XY coordinate plot.
Source Sec 9.6 XY scatter paragraph; objective Q1 (low scatter = good fit), Q6 (poor scatter = poor control), Q7 (fixed energy from XY coordinate system).
Source: BEE Guidebook 2014
📖 §9.6 Straight-line relationship y = c + mx (E = M·P + C)

22. Explain the energy-production straight-line relationship y = c + mx (E = c + mP). Define each term with units and state how the base load is found.

Model answer: The best-fit straight line through the energy-production scatter is y = c + mx, i.e. Energy consumed for the period = c + m x production for the same period. Here y is the dependent variable (energy consumption), x is the independent variable (production), c is the value where the line intersects the y-axis = the fixed energy consumption / base load (energy used even at zero production), and m is the gradient = the variable or specific energy consumption (extra energy per additional unit of production, e.g. toe/tonne). The base load c is found graphically as the y-intercept of the line on the XY coordinate plot. In the book's foundry example E = 180 + 0.4P, so the theoretical base load is 180 toe.
Source 'y = c + mx', definitions of y, x, c, m, and worked result E = 180 + 0.4x with base load 180 mtoe/toe.
Source: BEE Guidebook 2014
📖 §9.6 Normal equations / least squares (Example 9.1)

23. How is the best-fit straight line determined, and what are the 'normal equations' used to find c and m?

Model answer: The best-fit straight line is determined by the least-squares method - summing the squares of the distances of the data points from the line and minimising them. For a line y = c + mx fitted to n data points, the constants c and m are found from the two normal equations: c·n + m·Σx = Σy, and c·Σx + m·Σx² = Σxy, where n is the number of data points. Solving these simultaneously gives the slope m and intercept c (in the book's foundry example m = 0.4 and c = 180, giving y = 180 + 0.4x).
Source: 'cn + mΣx = Σy ; cΣx + mΣx² = Σxy ... known as the normal equations' and best-fit by summing squares of distances.
Source: BEE Guidebook 2014
📖 §9.6 Table 9.5 Minimum Correlation Coefficients

24. Why does the minimum acceptable correlation coefficient depend on the number of data samples?

Model answer: Although a best-fit line can always be drawn, with very scattered data the derived equation may be meaningless, so the reliability of the line (the correlation coefficient r) must meet a minimum acceptable value. The fewer the data points, the higher r must be to be considered reliable; as the number of samples rises, the minimum acceptable r falls. Per Table 9.5: 10 samples need r >= 0.767, 15 -> 0.641, 20 -> 0.561, 25 -> 0.506, 30 -> 0.464, 40 -> 0.402, 50 -> 0.362. The book's foundry example (9 points) with r = 0.98 is very good.
Source Table 9.5 'Minimum Correlation Coefficients (r)'.
Source: BEE Guidebook 2014
📖 §9.6 CUSUM — definition and purpose

25. What does CUSUM stand for and what is it used for in energy monitoring?

Model answer: CUSUM is an acronym for Cumulative Sum of differences. It is the cumulative summation, period by period, of the differences between actual energy consumption and the target/baseline (standard) consumption. CUSUM charts are particularly useful for diagnosing why excess energy is being consumed because they identify the date on which any change in energy performance occurred; knowing when a problem first occurred helps pinpoint it so that further analysis can find the root cause. Plotted against time, CUSUM reveals trends and lets energy savings or losses be quantified when performance changes.
Source: 'CUSUM is an acronym for Cumulative Sum of differences... they identify the date of any change in energy performance.'
Source: BEE Guidebook 2014
📖 §9.6 CUSUM — interpretation of line direction (Fig 9.12)

26. How is the direction of a CUSUM line interpreted?

Model answer: A typical CUSUM graph oscillates around the zero line (the baseline/standard) showing random fluctuation while performance is on-target. A change in direction indicates an event relevant to energy consumption. If the line goes UP, performance is worsening (specific energy consumption is going up - consuming more than predicted, due to poor control, housekeeping or maintenance). If the line goes DOWN, savings are being achieved (consuming less than predicted, e.g. after an energy-saving measure). A HORIZONTAL line means actual and calculated energy consumption are the same (on-target). Site knowledge is needed to interpret the events.
Source CUSUM paragraph; objective Q3 (going up -> SEC up) and Q10 (horizontal -> actual = calculated).
Source: BEE Guidebook 2014
📖 §9.6 CUSUM Example — Table 9.7 / Fig 9.13

27. In the foundry CUSUM example, how are the energy savings from the heat-recovery system read from the chart?

Model answer: Using the baseline equation E_calc = 0.4P + 180 from the first 9 months, the CUSUM oscillates around zero for several months and then drops sharply after month 11 - indicating the heat-recovery system took about two months to commission, after which steady savings were achieved. The savings equal the magnitude of the CUSUM drop: from -6 to -50 gives 50 - 6 = 44 toe accumulated over the last 7 months, which represents savings of almost 2% of energy consumption.
Source CUSUM example: 'savings of 44 toe (50-6) have been accumulated in the last 7 months... almost 2% of energy consumption.'
Source: BEE Guidebook 2014
📖 §9.6 Solved Example — CUSUM with E_calc = 0.5P + 220

28. An industry's baseline (Jan-Jun 2011) is E_calc = 0.5P + 220 (toe). After a waste-heat-recovery system, Jul-Dec data give a final CUSUM of -96. Find the energy saving and the reduction in specific energy consumption (Jul-Dec production = 4550 t).

Model answer: Energy saving = magnitude of the final CUSUM = 96 toe. Reduction in specific energy consumption = total saving / total production = 96 / 4550 = 0.021 toe/tonne of production. (Jul-Dec production = 760 + 820 + 940 + 750 + 610 + 670 = 4550 tonnes.)
Source solved example: 'Energy savings achieved = 96 toe; Reduction in SEC = 96/4550 = 0.021 toe/tonne'.
Source: BEE Guidebook 2014
📖 §9.7 EMIS — generic features

29. What is an Energy Management Information System (EMIS) and what generic features do EMIS software packages share?

Model answer: An EMIS is specially designed information-system software used to operate an M&T programme; computers are not a replacement for the energy manager but tools that store and analyse large amounts of data quickly. Generic features shared by EMIS packages: a database facility to store and organise large quantities of long-term data; ability to record energy data for all utility types from both meters and invoices; ability to handle complex utility tariffs; ability to handle related variables such as degree-days and production data; a statistical data-analysis facility; and a reporting facility that quickly produces energy management reports. Sophisticated packages can interface with Building Management Systems (BMS) to record data automatically (e.g. hourly).
Source Sec 9.7 EMIS generic features bullet list, including 'not a replacement for the energy manager' and BMS interface.
Source: BEE Guidebook 2014
📖 §9.7 Reporting by exception

30. What is 'reporting by exception' and why is it used in an M&T system?

Model answer: One disadvantage of producing many regular reports is that they swamp operational managers with apparently irrelevant information. Reporting by exception is a system in which reports are generated only when energy performance falls outside certain predetermined limits. Its advantages are that managers only receive reports when performance is either poor or very good, and everyone involved in the reporting process benefits from a reduced workload. Reports should be succinct, in a standard automatically-generated format, and published regularly so wasteful practices are identified quickly.
Source Sec 9.7 reporting-by-exception paragraph.
Source: BEE Guidebook 2014
📖 §9.7 Reporting frequency vs managerial status (Fig 9.14)

31. How should the frequency of energy management reports be matched to managerial level?

Model answer: Reports should be tailored to suit their readers, with different managers requiring different levels of report. The reporting frequency increases as the managerial level decreases: senior management typically needs only an annual or quarterly review; department heads monthly; and EAC (operational) managers weekly. Most M&T programmes publish reports weekly or monthly - monthly for large multi-site organisations and weekly (or even daily) for complex, high energy-consuming facilities. If the reporting period is too long, energy is wasted before action is taken; if too short, the system becomes over-complex with too much irrelevant information.
Source Sec 9.7 + Figure 9.14 (senior=quarterly/annual, dept head=monthly, EAC manager=weekly).
Source: BEE Guidebook 2014
📖 §9.6 Specific Energy Consumption charts (Figs 9.7-9.8)

32. What is Specific Energy Consumption (SEC) and what does the relationship between SEC and production reveal?

Model answer: Specific Energy Consumption (SEC) is the energy consumed per unit of production (e.g. toe/tonne or kWh/MT) and can be plotted as a bar chart against time. When production levels are added to the SEC chart, the features become clearer: a very low SEC occurs when production is at a record high, which indicates that there is a fixed (base-load) energy consumption - i.e. consumption that occurs regardless of production level. This is why energy consumption mostly relates to production, and SEC must be interpreted together with output.
Source Sec 9.6 SEC charts (Figures 9.7/9.8): low SEC at record production indicates fixed energy consumption.
Source: BEE Guidebook 2014
📖 §9.6 Table 9.4 Factors which influence Energy Consumption

33. Give examples of factors that influence energy/water consumption for different end-uses.

Model answer: Regression depends on choosing the right influencing variable for each end-use. Examples from the guidebook (Table 9.4): for electricity used by air compressors the influencing factor is the air volume delivered; for furnace oil used for steam raising in boilers it is the amount of steam generated; for steam used in a production process it is the production volume. In general, the variables commonly compared in energy regression are fuel/electricity/water consumption versus units of production, and lighting electricity versus hours of occupancy.
Source Table 9.4 'Factors which influence Energy Consumption' and the regression variable examples.
Source: BEE Guidebook 2014
📖 §9.6 Influencing / normalising variables (Book EOC Q5 & Q9)

34. To what does a company's energy consumption mostly relate, and which type of factor does NOT contribute to it?

Model answer: For most companies, energy consumption mostly relates to production. Genuine influencing/normalising factors include production (volume/output), operating hours, and climate (degree-days/weather). Factors that do not genuinely drive energy use - such as profits, inventory, or maintenance cost - are not used as normalising variables. Choosing the correct driving variable is essential, since regression is only meaningful when the chosen variable actually influences energy use.
Source objective Q9 (energy relates to production) and Q5 (which variable does not contribute); notes flag maintenance cost as a non-normalising-factor trap.
Source: BEE Guidebook 2014
📖 §9.6 Annual energy consumption using bar chart (Fig 9.3)

35. How is annual energy consumption represented using a bar chart, and what is its limitation?

Model answer: If 24 months of energy data are collated, annual energy consumption can be shown as a bar chart. The most common application in energy management plots energy per month for the current year against the previous year, allowing month-by-month comparison. Its limitation is that this chart does not clearly tell us about any trends in energy consumption - it only compares two years side by side without normalising for production or other variables.
Source Sec 9.6 'Annual Energy Consumption Using Bar Chart' (Figure 9.3).
Source: BEE Guidebook 2014
📖 §9.6 Graphical correlation of production & energy (Book EOC Q8)

36. What is the best way of correlating production and energy data in a plant, and why?

Model answer: The best way of correlating production and energy data is graphical representation - specifically an XY scatter plot of energy versus production with a best-fit straight line. Graphical representation makes the linear relationship E = c + mP visible, lets the fixed (base-load) energy be read as the y-intercept and the variable/specific energy as the slope, and shows the degree of scatter (hence the level of control). Text format or oral communication cannot reveal these relationships.
Source objective Q8 (best way = graphical representation), supported by the XY scatter section.
Source: BEE Guidebook 2014
📖 §9.6 Plant Energy Performance & production factor (M&T normalisation)

37. Calculate the production factor and plant energy performance and comment. Reference year: energy 10 million kcal, production 90,000 MT. Current year: energy 8 million kcal, production 70,000 MT.

Model answer: Production factor = current-year production / reference-year production = 70,000 / 90,000 = 0.778. Reference-year equivalent energy = reference-year energy x production factor = 10 x 0.778 = 7.78 million kcal (the energy the plant SHOULD have used at the current output). Plant Energy Performance (PEP) = (Reference-year equivalent - Current-year energy) / Reference-year equivalent x 100 = (7.78 - 8.00) / 7.78 x 100 = -0.222/7.78 x 100 = -2.86% (about -2.9%). COMMENT: PEP is NEGATIVE, so plant energy performance has WORSENED. Normalised to the lower current output the plant should have needed only 7.78 million kcal but actually consumed 8 million kcal - roughly 2.9% more than the production-normalised reference. A POSITIVE PEP would have indicated improvement.
Method matches Ch9; the production-factor/plant-energy-performance procedure is not in the OCR body so verified=false (method is consistent with the guidebook approach taught for Ch9). Corrected: the production factor 70,000/90,000 = 0.7778 must be carried through, giving PEP = -2.86% (not -2.56% from a rounded 0.78).
Source: Mar 2023 Paper-1 (23rd National Certification Exam)
📖 §9.6 Fixed energy from E = M·P + C

38. Calculate the fixed energy consumption for a rolling mill consuming 1,00,000 units of electricity to produce 600 MT/month with a specific energy consumption of 100 kWh/MT.

Model answer: Total energy = Fixed energy + (SEC x production), i.e. E = c + mP. Therefore fixed energy c = Total - (SEC x production) = 1,00,000 - (100 x 600) = 1,00,000 - 60,000 = 40,000 units. The variable (production-related) energy is 60,000 units and the fixed/base-load energy is 40,000 units.
Applies the Ch9 relation E = c + mP. Past-paper numeric, not in OCR body, so verified=false.
Source: Jul 2010 Paper-1 Set A (10th National Certification Exam)
📖 §9.2/§9.6 Normalising of data & benchmarking

39. What is meant by (a) Normalising of data and (b) Benchmarking?

Model answer: (a) Normalising of data is the process of removing the impact of variable factors (such as weather/degree-days, production level, occupancy, operating characteristics) on energy use so that energy performance can be compared on a like-for-like basis and a meaningful baseline established. (Note: maintenance cost is NOT a normalising factor.) (b) Benchmarking is the comparison of one's energy performance against that of peers, competitors or similar organisations to establish a relative understanding of where one's performance ranks, and to set realistic external targets.
Concept consistent with Ch9 (targeting uses benchmarking with similar organisations); definitions are standard. Not verbatim in OCR -> verified=false.
Source: Jul 2010 Paper-1 Set A (10th National Certification Exam)
📖 §9.6 Example 9.1 — E = 0.4P + 180 (m and c with units)

40. Given the energy-production relationship E = 0.4P + 180 (E in toe/month, P in tonnes/month), identify the fixed and variable energy consumption with units.

Model answer: Comparing with E = c + mP: the intercept c = 180 toe/month is the fixed energy consumption (theoretical base load) - the energy used even at zero production. The slope m = 0.4 toe/tonne is the variable or specific energy consumption - the extra energy required per additional tonne of production. So at, say, 500 tonnes/month the predicted energy is 0.4 x 500 + 180 = 380 toe/month.
Directly applies the OCR foundry result 'y = 180 + 0.4x... theoretical base load for furnace is 180 mtoe'. Worked interpretation, not verbatim Q -> verified=false.
Source: BEE Guidebook 2014
📖 §9.7 EMIS — role of computers in M&T

41. What is the role of computers/software in an M&T programme, and why are they not a replacement for the energy manager?

Model answer: Specially designed information-system software (EMIS) is advisable for operating an M&T programme because it can store and analyse large amounts of data in a short period. Its great advantage is the database facility, which lets historical data and data from many sources be instantly compared - useful for comparing site energy costs and quickly assessing the relative performance of EACs, so under-performing EACs can be identified and remedial action taken. However, computers should not be seen as a replacement for the energy manager but simply as tools; interpretation of results and decisions still require the manager's judgement and site knowledge.
Source Sec 9.7: 'Computers should not be seen as a replacement for the energy manager, but simply as tools...'. Synthesised short Q -> verified=false.
Source: BEE Guidebook 2014
📖 §9.3 Classification of organisations & assignment of EACs

42. How are organisations classified from an energy point of view when designing an M&T programme, and how are EACs assigned in each case?

Model answer: Organisations are typically classified by the number of sites and the level of metering: (1) single site with central utility metering - best treated as a single EAC; (2) single site with sub-metering - the site can be broken up into several separate EACs; (3) multi-site with central utility metering - each site treated as a separate EAC; (4) multiple sites with sub-metering - each site can be divided into several separate EACs. The M&T programme must be designed to suit the needs of the particular organisation.
Source Sec 9.3 four classifications and EAC assignment. Synthesised into a short Q (content verbatim) -> verified true on content but framed as new Q; marking verified=true since fully grounded in OCR.
Source: BEE Guidebook 2014
📖 Book-1 §9.6 Data and Information Analysis — time-dependent analysis, norm chart, deviance chart, MAT, XY scatter

43. Briefly compare the main graphical tools used in M&T: time-dependent graph, norm chart, deviance chart, MAT and XY scatter.

Model answer: Time-dependent graph (energy vs time): shows trends and seasonal patterns but cannot explain why or confirm a trend - comparative only. Norm chart: actual overlaid on target; weak analytically but easy for managers to understand and good for highlighting exceptions. Deviance chart: plots (actual - target), positive above target/negative below; excellent for flagging problems and triggering exception reports. Moving Annual Total (MAT): each point sums the previous 12 months, removing seasonal effects and meter-timing errors. XY scatter: energy vs production, revealing the linear relationship, the fixed load (y-intercept) and the level of control (scatter).
Confirmed vs Book-1 §9.6 — each description is taken from the guidebook: the time-dependent graph "can be used only as a comparative tool and not an absolute one"; the norm chart "is of little value as an analytical tool, but can be useful for highlighting exceptions"; deviance charts "are particularly good at highlighting problems" and can initiate exception reports. MAT sums the previous 12 months, removing seasonal effects and meter-timing errors; the XY scatter reveals the E = c + mP relationship, the base load (intercept) and the degree of control (scatter). Consolidated for revision, but every statement is traceable to §9.6.
Source: BEE Guidebook 2014
📖 Book-1 §8.8 Measurement and Verification (also §7.8 Calculating savings)

44. What is meant by Measurement & Verification (M&V) of energy savings, and why are adjustments needed?

Model answer: Measurement & Verification (M&V) is the process of quantifying the energy savings from an efficiency measure by comparing energy use before and after implementation. Savings = base-year (pre-retrofit) energy - post-retrofit energy +/- adjustments. Adjustments are needed to normalise for changes in factors that affect energy independently of the retrofit - such as production level, weather/degree-days and occupancy - so that the reported saving reflects only the effect of the measure and not changes in operating conditions. This is the same baseline/normalisation principle used in M&T.
Corrected — Book-1 §8.8: the previous note claimed M&V "is not in the OCR body". It is in the guidebook: §8.8 defines M&V as "a process which is used to determine energy and demand savings" and gives Energy Savings = Base year Energy Use − Post-Retrofit Energy Use ± Adjustments; §7.8 repeats it as Baseline − Current ± Adjustment. Adjustments bring the two periods to the same set of conditions — weather, occupancy, plant throughput and equipment operations — and are derived from identifiable physical facts. M&V matters most where contract payments or a performance guarantee depend on the magnitude of the savings.
Source: BEE practice (M&V concept linked to Ch9 baseline)

Long questions (10 marks) — 16

📖 §9.1-§9.4, §9.7 — M&T system: definition, EACs, six key elements, benefits, reporting by exception

1. Describe an Energy Monitoring and Targeting (M&T) system. Define M&T, distinguish monitoring from targeting, explain the role of Energy Account Centres (EACs), list and explain the six key elements of an M&T system, state its benefits, and explain 'reporting by exception'. (10 marks)

Model answer: DEFINITION: Energy Monitoring & Targeting (M&T) is primarily a MANAGEMENT technique that uses energy information as the basis to eliminate waste, reduce and control the current level of energy use, and improve existing operating procedures. It is based on the principle "you can't manage what you don't measure" and combines the principles of energy use with statistics. Typical M&T programmes reduce annual energy costs by 5-15%. MONITORING vs TARGETING: - Monitoring = the process of establishing the existing pattern of energy consumption and explaining deviations from that pattern (its goal is to maintain the existing pattern by providing data on energy consumption and related data such as production). - Targeting = the identification of a desirable (lower) energy consumption level and working towards achieving it. Targets are based on historical (average or best) data from monitoring plus benchmarking against similar organisations. ENERGY ACCOUNT CENTRES (EACs): Before initiating M&T, set up EACs - departments, processes or cost centres. Operational managers are made accountable for the energy consumption of the EAC(s) they control. Sub-metering allows a single site to be split into several EACs; multi-site organisations treat each site (or each sub-metered area) as a separate EAC. SIX KEY ELEMENTS OF AN M&T SYSTEM (in order): 1. Recording - measuring and recording the energy consumption of each EAC (set up procedures for regular, reliable data collection). 2. Analysing & Comparing - relating energy consumption to a measured output (e.g. production) using 12-24 months of historical data to obtain the Standard Energy Performance for each EAC (via regression). This standard is the BASELINE and can also serve as the initial target. 3. Setting Targets - setting achievable targets that improve on the standard energy performance, based on benchmarking or the least historical consumption achieved. 4. Monitoring - comparing actual energy consumption against the set target on a regular basis. 5. Reporting - reporting results to management, including variances from targets and related equipment/system performance problems. Reports stimulate improvement and quantify achievements. 6. Controlling - implementing management measures to correct any variances that occur. An M&T system also: checks the accuracy of energy invoices; allocates energy costs to EACs; determines energy performance/efficiency; records energy use so efficiency projects can be verified; and highlights performance problems in equipment/systems. BENEFITS: reduce energy cost through improved efficiency; identify and explain increases/decreases in use; draw consumption trends (weekly/seasonal/operational); improve energy budgeting to match production plans; observe reactions to past changes; forecast future use when planning operational changes; diagnose areas of wasted energy; develop performance targets; and manage energy as a controllable resource rather than a fixed uncontrollable cost. REPORTING BY EXCEPTION: a system in which reports are generated ONLY when energy performance falls outside predetermined limits. This avoids swamping managers with routine/irrelevant reports - they receive a report only when performance is either poor or exceptionally good, reducing everyone's workload.
High-frequency 10-mark theory long. Memorise the 6 elements in order (Record -> Analyse & Compare -> Set Targets -> Monitor -> Report -> Control) and the principle 'you can't manage what you don't measure'. Anchor figures: 5-15% cost reduction; 12-24 months of data for the baseline; EACs set up FIRST. Reporting frequency rises with operational level (senior mgmt = quarterly/annual, EAC manager = weekly). EMIS software supports M&T but is NOT a replacement for the energy manager.
Source: unknown
📖 §9.6 CUSUM technique (Fig 9.12, Table 9.7)

2. Explain the CUSUM (Cumulative Sum of differences) technique used in Energy Monitoring & Targeting. State what CUSUM is, why it is useful, the standard energy performance equation it uses, the step-by-step procedure to build a CUSUM table, and how the direction of the CUSUM line is interpreted. (10 marks)

Model answer: WHAT IT IS: CUSUM is an acronym for Cumulative Sum of differences. It is the running (cumulative) total, period by period, of the difference between ACTUAL energy consumption and the STANDARD (predicted/baseline) energy consumption. WHY IT IS USEFUL: CUSUM charts are particularly useful for diagnosing WHY and especially WHEN a change in energy performance occurred - they pinpoint the DATE of any change in performance (the point where the line changes direction). Knowing when a problem first occurred helps pin-point its root cause, and CUSUM also lets you quantify the energy savings or losses when performance changes. BASELINE EQUATION: First establish the standard energy performance equation from data (energy vs a related variable such as production) collected during a monitoring period BEFORE any intervention. This is the straight line E_calc = m.P + c (book form y = c + mx), where m = variable/specific consumption (slope) and c = fixed/base-load consumption (intercept). This equation gives the target/baseline against which actual consumption is compared. STEPS FOR CUSUM ANALYSIS: 1. Plot the Energy vs Production graph for the pre-intervention months. 2. Draw the best-fit straight line through the points. 3. Derive the equation of the line, E = m.P + c (e.g. book: E = 0.4P + 180). 4. Calculate the standard/predicted consumption E_calc for each month from the equation, using that month's production. 5. Calculate the difference for each month: diff = E_actual - E_calc. 6. Compute CUSUM = the cumulative (running) sum of these differences. 7. Plot the CUSUM graph against time. 8. Estimate the savings accumulated (the size of the CUSUM drop). INTERPRETING THE LINE DIRECTION: - A typical CUSUM graph oscillates randomly around zero (the baseline) while performance is unchanged. This continues until 'something happens' (an energy-saving measure, or a worsening due to poor control/housekeeping/maintenance). - Line going DOWN = actual is below predicted = SAVINGS (energy consumption reduced). - Line going UP = actual above predicted = performance WORSENING (specific energy consumption rising). - Line HORIZONTAL / oscillating about zero = actual equals calculated (on-target, no change). - The MAGNITUDE of the drop between the point the line starts to fall and the end of the period = the energy saving. Site knowledge is needed to correctly attribute each change in direction to a real event.
This is the concept half of the guaranteed CUSUM question - the numerical half (build the table) always accompanies it. Objective traps drawn from this: CUSUM line UP = SEC going up (worsening); horizontal = actual = calculated. CUSUM's unique advantage over other tools is that it pinpoints the DATE performance changed. Savings = (CUSUM at start of the fall) - (CUSUM at end).
Source: unknown
📖 §9.6 Linear regression, normal equations & Pearson correlation coefficient (Table 9.5)

3. Explain linear regression analysis as an energy management tool. Give the straight-line energy-production relationship and the meaning of each term, state the least-squares normal equations used to find the constants, and explain the Pearson correlation coefficient and its significance. (10 marks)

Model answer: PURPOSE: Linear regression analysis is a statistical technique that determines and quantifies the relationship between variables. It enables standard energy-consumption equations to be established from data that would otherwise be meaningless. It overcomes the limitation of time-dependent analysis by removing the 'time' element and focusing on the variables that actually influence energy consumption (e.g. furnace-oil vs units produced, electricity vs production, water vs production, lighting electricity vs occupancy hours). THE STRAIGHT-LINE RELATIONSHIP: The best-fit line is found by minimising the sum of the squares of the distances of the data points from the line (least squares). Its generic equation is: y = c + m.x i.e. Energy for the period = c + m x (Production for the same period) where: - y = dependent variable = energy consumption - x = independent variable = production - c = intercept = the value where the line cuts the y-axis = FIXED (base-load) energy consumption, used even at zero production - m = gradient/slope = VARIABLE (specific) energy consumption per unit of production Once established, this equation predicts future consumption and serves as the standard performance equation for M&T. Fixed energy consumption is therefore read from the XY-coordinate (y-intercept), not from a bar/pie chart. NORMAL EQUATIONS (to find c and m by hand, n = number of data points): n.c + m.(Sum x) = (Sum y) c.(Sum x) + m.(Sum x^2) = (Sum xy) Solving these two simultaneous equations gives c and m. PEARSON CORRELATION COEFFICIENT (r): Because scattered data can give a meaningless line, r tests how well the best-fit line correlates with the sample data - i.e. the reliability of the line. r lies between 0 and 1, where 1 represents 100% (perfect) correlation. It is computed from the deviations of x and y from their means: r = Sum[(x - x_mean)(y - y_mean)] / sqrt{ Sum[(x - x_mean)^2] . Sum[(y - y_mean)^2] } The more data points there are, the lower the minimum acceptable r (e.g. 10 samples -> 0.767; 20 -> 0.561; 30 -> 0.464). A low degree of scatter (high r) = good fit = good control; a poor fit (low r), where a relationship is expected, indicates poor control and hence potential/scope for energy savings. MULTI-VARIABLE CASE: When several variables influence energy, y = c + m1.x1 + m2.x2 + ... + mn.xn. This is difficult by hand and is best solved with specialist computer software.
Underpins every regression/CUSUM numerical. Remember: m = variable/specific (slope), c = fixed/base load (intercept read from y-axis). r=1 is perfect; book foundry example gives r=0.98 (very good). Poor scatter = poor control = savings potential is a recurring objective. The two normal equations are worth memorising verbatim - they are the exact tool for the 'find the equation first' numericals.
Source: unknown
📖 §9.6-§9.7 Graphical tools (pie/bar/time/norm/deviance/MAT/XY scatter) and EMIS

4. Describe the graphical and analytical tools used in Energy Monitoring & Targeting for identifying trends and projections: annual consumption breakdown, time-dependent graph, norm chart, deviance chart, moving annual total, XY scatter diagram, and the Energy Management Information System (EMIS). (10 marks)

Model answer: 1. ANNUAL ENERGY CONSUMPTION BREAKDOWN: Convert all fuels/energy to standard units (kcal, e.g. electricity 860 kcal/kWh, HSD 10,500, furnace oil 10,200, LPG 12,000 kcal/kg), tabulate annual consumption and cost, then produce a percentage breakdown and pie-charts of energy and cost share. Quickly shows overall performance and trends, but makes no allowance for variable factors (climate, occupancy) so it cannot be used to compare different organisations. 2. BAR CHART (current vs previous year): 24 months of data plotted as monthly bars for two years; shows level but does not clearly reveal consumption trends. 3. TIME-DEPENDENT GRAPH (energy vs time): Reveals general trends and seasonal patterns and highlights exceptions to the norm. Limitation: it cannot explain WHY a trend occurs; it is a comparative, not an absolute, tool. More than one variable (e.g. oil + electricity) can be plotted. 4. NORM CHART: A sequential plot of ACTUAL consumption overlaid on TARGET consumption. Of little analytical value, but good for highlighting exceptions and easy for senior/operational managers to understand (a historical record). 5. DEVIANCE CHART: Plots the DIFFERENCE (target - actual). Consumption above target is plotted positive, below target negative; showing upper/lower normal-operation limits distinguishes normal variation from serious deviation. Excellent for highlighting problems and triggering exception reports. 6. MOVING ANNUAL TOTAL (MAT): Each point = the sum of the previous 12 months of data. Every point therefore spans a full range of seasons/holidays and the technique smooths out meter-reading timing errors. If energy and production MATs track each other there is no cause for alarm; divergence of the energy line gives early warning of waste or confirms savings. 7. XY SCATTER DIAGRAM: Plots energy (dependent) against production (independent) to reveal their relationship. A low degree of scatter = good fit = good control; poor scatter (where a relationship is expected) = poor control = scope for savings. The best-fit line yields the standard equation and the fixed load (y-intercept). 8. EMIS (Energy Management Information System): Purpose-designed software to store and analyse large amounts of M&T data. Generic features: a database facility; recording of all utility types from meters and invoices; handling of complex tariffs; handling of related variables (degree-days, production); a statistical data-analysis facility; and a reporting facility. Advanced packages can interface with Building Management Systems (BMS) for automatic (e.g. hourly) data capture. EMIS is a TOOL - not a replacement for the energy manager.
A comprehensive 'describe the tools' 10-marker. Key discriminators for objectives: fixed energy is found from the XY-coordinate (scatter) y-intercept; poor scatter = poor control = savings potential; MAT smooths seasonal/meter-timing errors; deviance chart plots (target - actual); EMIS is not a replacement for the energy manager. The standard energy conversions (860/10,500/10,200/12,000 kcal) are worth memorising.
Source: unknown
📖 §9.6 Example 9.1 — foundry least-squares regression (E = 0.4P + 180, r = 0.98)

5. A foundry monitoring programme produced the following monthly data - Production X (tonnes): 380, 440, 460, 520, 320, 520, 240, 620, 600; Energy use Y (toe): 340, 340, 380, 380, 300, 400, 280, 424, 420. Using the least-squares normal equations, derive the best-fit energy-production equation, state the base (fixed) load, and comment on the fit using the correlation coefficient. (10 marks)

Model answer: STEP 1 - Build the sums table (n = 9): Month | X | Y | X^2 | XY 1 | 380 | 340 | 144400 | 129200 2 | 440 | 340 | 193600 | 149600 3 | 460 | 380 | 211600 | 174800 4 | 520 | 380 | 270400 | 197600 5 | 320 | 300 | 102400 | 96000 6 | 520 | 400 | 270400 | 208000 7 | 240 | 280 | 57600 | 67200 8 | 620 | 424 | 384400 | 262880 9 | 600 | 420 | 360000 | 252000 Totals: Sum X = 4100, Sum Y = 3264, Sum X^2 = 1,994,800, Sum XY = 1,537,280. STEP 2 - Write the normal equations (n.c + m.SumX = SumY ; c.SumX + m.SumX^2 = SumXY): 9c + 4100m = 3264 ...(i) 4100c + 1,994,800m = 1,537,280 ...(ii) STEP 3 - Solve. From (i): c = (3264 - 4100m)/9. Substitute into (ii): 4100(3264 - 4100m)/9 + 1,994,800m = 1,537,280 1,486,933 - 1,867,778m + 1,994,800m = 1,537,280 127,022 m = 50,347 => m = 0.4 c = (3264 - 4100 x 0.4)/9 = (3264 - 1640)/9 = 180 STEP 4 - Best-fit equation: E = 0.4 P + 180 (toe/month) - Variable/specific consumption m = 0.4 toe per tonne of production. - Fixed (base-load) consumption c = 180 toe/month (theoretical base load of the furnace at zero production). STEP 5 - Correlation coefficient: using the deviations from the means, Sum[(x-xbar)(y-ybar)] = 50,346.67, Sum[(x-xbar)^2] = 127,022.22, Sum[(y-ybar)^2] = 20,832. r = 50,346.67 / sqrt(127,022.22 x 20,832) = 0.98 Since r = 0.98 (close to 1) and, for 9-10 samples, the minimum acceptable r is about 0.767, the correlation is very good - the line is a reliable predictor of energy consumption and can be used as the standard performance equation for M&T.
The canonical worked regression. Memorise the two-normal-equation drill: build Sum X, Sum Y, Sum X^2, Sum XY; solve the 2x2; read m (variable) and c (fixed = base load). Result E = 0.4P + 180 is reused as the baseline in the 18-month CUSUM example. r = 0.98 shows an excellent fit.
Source: unknown
📖 §9.6 Solved Example — CUSUM with E_calc = 0.5P + 220 (96 toe saving)

6. The energy-production data of an industry (Jan-June 2011) follows the baseline relationship: Calculated energy consumption E_calc = 0.5 P + 220 (toe/month). A waste-heat-recovery system was installed at the end of June 2011 and further data gathered Jul-Dec 2011. Monthly actual energy (toe) and production (tonnes) were - Jul 590/760, Aug 605/820, Sep 670/940, Oct 582/750, Nov 512/610, Dec 540/670. Using the CUSUM technique, calculate the energy savings (toe) and the reduction in specific energy consumption. (10 marks)

Model answer: Compute E_calc = 0.5P + 220 for each post-intervention month, then diff = E_act - E_calc, then the running CUSUM: Month | E_act (toe) | P (t) | E_calc = 0.5P+220 | E_act - E_calc | CUSUM Jul | 590 | 760 | 600 | -10 | -10 Aug | 605 | 820 | 630 | -25 | -35 Sep | 670 | 940 | 690 | -20 | -55 Oct | 582 | 750 | 595 | -13 | -68 Nov | 512 | 610 | 525 | -13 | -81 Dec | 540 | 670 | 555 | -15 | -96 The CUSUM trends steadily DOWN (actual below predicted every month), confirming the waste-heat-recovery system is delivering steady savings. ENERGY SAVINGS = magnitude of the final CUSUM = 96 toe (over the 6 months Jul-Dec 2011). REDUCTION IN SPECIFIC ENERGY CONSUMPTION: Total production Jul-Dec = 760 + 820 + 940 + 750 + 610 + 670 = 4550 tonnes. SEC reduction = total savings / total production = 96 / 4550 = 0.021 toe/tonne of production.
This is the book's own solved example (p.233) and a past 10-mark long - it is the template for the guaranteed CUSUM question. Note the two-part deliverable: (a) savings = final CUSUM magnitude = 96 toe; (b) SEC reduction = savings / total production = 96/4550 = 0.021 toe/t. Down-trending CUSUM = savings.
Source: unknown
📖 §9.6 CUSUM Example — Table 9.7 (18-month foundry, 44 toe saving)

7. Energy and production data were collected for a foundry furnace over 18 months; a heat-recovery system was installed in month 9. The pre-intervention regression gives the baseline E = 0.4 P + 180 (toe). Given monthly actual energy E_act (toe) and production P (tonnes) - M1 340/380, M2 340/440, M3 380/460, M4 380/520, M5 300/320, M6 400/520, M7 280/240, M8 424/620, M9 420/600, M10 400/560, M11 360/440, M12 320/360, M13 340/420, M14 372/480, M15 380/540, M16 280/280, M17 280/260, M18 380/500 - construct the CUSUM table and estimate the savings from the heat-recovery system. (10 marks)

Model answer: For each month compute E_calc = 0.4P + 180, then diff = E_act - E_calc, then the running CUSUM (as in Guide Book Table 9.7): Month | E_act | P | E_calc=0.4P+180 | E_act-E_calc | CUSUM 1 | 340 | 380 | 332 | +8 | +8 2 | 340 | 440 | 356 | -16 | -8 3 | 380 | 460 | 364 | +16 | +8 4 | 380 | 520 | 388 | -8 | 0 5 | 300 | 320 | 308 | -8 | -8 6 | 400 | 520 | 388 | +2 | -6 7 | 280 | 240 | 276 | +4 | -2 8 | 424 | 620 | 428 | -4 | -6 9 | 420 | 600 | 420 | 0 | -6 10 | 400 | 560 | 404 | -4 | -10 11 | 360 | 440 | 356 | +4 | -6 12 | 320 | 360 | 324 | -4 | -10 13 | 340 | 420 | 348 | -8 | -18 14 | 372 | 480 | 372 | 0 | -18 15 | 380 | 540 | 396 | -16 | -34 16 | 280 | 280 | 292 | -12 | -46 17 | 280 | 260 | 284 | -4 | -50 18 | 380 | 500 | 380 | 0 | -50 INTERPRETATION: The CUSUM oscillates around zero for the first several months, then drops sharply and steadily from about month 11 onwards. This shows the heat-recovery system took roughly two months to commission and reach proper operating conditions, after which steady savings were achieved. SAVINGS = the size of the CUSUM drop = from about -6 (month 9/11, before the fall) to -50 (month 18) = 50 - 6 = 44 toe accumulated over the last 7 months, which is almost 2% of energy consumption.
The second book worked CUSUM example. It teaches interpretation as well as arithmetic: the flat-then-falling shape shows commissioning lag (~2 months) before steady savings, and the saving is the vertical drop of the CUSUM (50-6 = 44 toe, ~2%). Reproduce the book's Table 9.7 values to match the key; note the single month-6 arithmetic quirk in the printed table. In the printed Guide Book Table 9.7 the month-6 difference is shown as +2 (giving CUSUM -6), which is what the book's savings figure of 44 toe is based on; strictly, E_act 400 - E_calc 388 = +12. The table above reproduces the book values so the marking key (44 toe) is matched.
Source: unknown
📖 §9.6 CUSUM — Book end-of-chapter Long Question L-1

8. In a food processing plant the monthly production-related (variable) energy consumption was 1.8 times the production and the non-production-related (fixed) energy consumption was 15,000 kWh/month up to May 2010. From June 2010 a series of energy conservation measures were implemented. Using the CUSUM technique, develop a table and calculate the energy savings for the subsequent 6 months from the data - Jul 62000 kg/113600 kWh, Aug 71000/139000, Sep 75000/158000, Oct 59000/119300, Nov 62000/123700, Dec 73000/143600. (10 marks)

Model answer: STEP 1 - Baseline (pre-June-2010) equation: variable = 1.8 x production, fixed = 15,000 kWh/month, so E_calc = 1.8 P + 15,000 (kWh, P in kg) STEP 2 - For each month compute E_calc, diff = E_act - E_calc, and the running CUSUM: Month | P (kg) | E_act (kWh) | E_calc = 1.8P+15000 | E_act-E_calc | CUSUM Jul | 62000 | 113600 | 126600 | -13000 | -13000 Aug | 71000 | 139000 | 142800 | -3800 | -16800 Sep | 75000 | 158000 | 150000 | +8000 | -8800 Oct | 59000 | 119300 | 121200 | -1900 | -10700 Nov | 62000 | 123700 | 126600 | -2900 | -13600 Dec | 73000 | 143600 | 146400 | -2800 | -16400 STEP 3 - Interpretation and result: The net CUSUM is negative, i.e. actual consumption is below the pre-June baseline overall, confirming the conservation measures are saving energy (September was the only month slightly above baseline). ENERGY SAVINGS over Jul-Dec 2010 = magnitude of the final CUSUM = 16,400 kWh.
A book end-of-chapter long (L-1) and a 'find-the-equation-first' CUSUM: the baseline E = 1.8P + 15000 must be assembled from the words (variable coefficient 1.8, fixed 15,000) before the table. Total saving = final CUSUM magnitude = 16,400 kWh. Watch units (P in kg, energy in kWh).
Source: unknown
📖 §9.6 CUSUM in SEC form — Book end-of-chapter Long Question L-2

9. A plant implemented energy saving measures prior to January 2011. Using the CUSUM technique, calculate the energy savings for the first 6 months of 2011. Average production Jan-Jun 2011 is 1000 MT/month. Actual and Predicted specific energy consumption (kWh/MT) are - Jan 1203/1121, Feb 1187/1278, Mar 1401/1571, Apr 1450/1550, May 1324/1284, Jun 1233/1233. (10 marks)

Model answer: STEP 1 - Because the data is specific energy consumption (SEC), the per-month deviation is diff = Actual SEC - Predicted SEC (kWh/MT); a NEGATIVE diff = saving. Build the running CUSUM: Month | Actual SEC | Predicted SEC | Actual-Predicted | CUSUM (kWh/MT) Jan | 1203 | 1121 | +82 | +82 Feb | 1187 | 1278 | -91 | -9 Mar | 1401 | 1571 | -170 | -179 Apr | 1450 | 1550 | -100 | -279 May | 1324 | 1284 | +40 | -239 Jun | 1233 | 1233 | 0 | -239 STEP 2 - Interpretation: after an adverse January, the CUSUM falls strongly (Feb-Apr) then flattens, giving a net cumulative SEC deviation of -239 kWh/MT, i.e. actual SEC is below predicted overall = savings. STEP 3 - Total energy savings: with average production 1000 MT/month, total energy saved = |net CUSUM| x average monthly production = 239 kWh/MT x 1000 MT = 239,000 kWh (approx. 2,39,000 kWh) over Jan-Jun 2011.
Book end-of-chapter long (L-2). The twist: data is given directly as Actual vs Predicted SEC, so no baseline equation is needed - just diff = Actual - Predicted and the running sum. Convert the cumulative SEC saving to energy by multiplying by production: 239 kWh/MT x 1000 MT = 239,000 kWh. Net negative CUSUM = saving.
Source: unknown
📖 §9.6 Plant Energy Performance & production factor (M&T normalisation; PAT context)

10. An integrated paper plant produced 119,366 MT of paper during 2012-13 (reference year) at a specific energy consumption of 53 GJ/tonne. Energy conservation measures under the PAT scheme reduced the SEC to 50 GJ/tonne. Actual production in the assessment year (2014-15) was 124,141 MT. Calculate the plant energy performance and state your inference. (10 marks)

Model answer: Reference year (2012-13): production = 119,366 MT; SEC = 53 GJ/tonne. Assessment year (2014-15): production = 124,141 MT; SEC = 50 GJ/tonne. STEP 1 - Production Factor = Assessment-year production / Reference-year production PF = 124,141 / 119,366 = 1.04 STEP 2 - Reference-year energy use = 53 x 119,366 = 6,326,398 GJ STEP 3 - Assessment-year (actual) energy use = 50 x 124,141 = 6,207,050 GJ STEP 4 - Reference-year-equivalent energy (energy that WOULD have been used at the assessment-year output) = Reference-year energy x Production Factor = 6,326,398 x 1.04 = 6,579,454 GJ STEP 5 - Plant Energy Performance = (Ref-equivalent energy - Actual energy) / Ref-equivalent energy x 100 = (6,579,454 - 6,207,050) / 6,579,454 x 100 = 372,404 / 6,579,454 x 100 = 5.66% INFERENCE: The plant energy performance is POSITIVE (+5.66%), meaning the plant used 5.66% LESS energy than the production-normalised reference - i.e. the plant is achieving genuine energy savings after the conservation measures.
Standard Plant Energy Performance / Production Factor numerical (a frequent M&T short/long). Method: PF = current/reference production; normalise the reference energy by PF; performance % = (ref-equivalent - actual)/ref-equivalent x 100. POSITIVE = improvement/savings; NEGATIVE = worse. Always state the sign-based inference.
Source: unknown
📖 §9.6 Plant Energy Performance & production factor + MTOE data for designated consumer

11. (a) A round-the-clock manufacturing plant with large heating/cooling needs uses grid electricity, furnace oil (thermic fluid heaters), coal (steam boilers), HSD (diesel generators) and LPG (ovens). To assess whether the plant qualifies as a designated consumer under the EC Act, list the data required to compute the MTOE. (b) Reference year (2022) energy use = 20 million kcal; production factor for the current year (2023) = 0.9; current year energy use = 19 million kcal. Calculate the plant energy performance for 2023 and state the inference. (10 marks)

Model answer: (a) DATA REQUIRED (each energy source, its use, and its metering unit - each is converted to a common toe/MTOE basis using its calorific value): - Electricity - purchased from the grid - in kWh - Furnace Oil - thermic fluid heaters - in litres (or kg) - Coal - steam boilers - in metric tonnes - HSD (High Speed Diesel) - diesel generators - in litres - LPG - ovens - in kilograms Each quantity is multiplied by its gross calorific value and summed to obtain the total energy in a common unit (kcal -> toe -> MTOE); comparing the total MTOE against the EC Act threshold determines designated-consumer status. (b) PLANT ENERGY PERFORMANCE for 2023: Reference-year-equivalent energy = Reference-year energy x Production Factor = 20 x 0.9 = 18 million kcal. Plant Energy Performance = (Ref-equivalent - Current-year energy) / Ref-equivalent x 100 = (18 - 19) / 18 x 100 = -5.56% INFERENCE: The performance is NEGATIVE (-5.56%), so performance WORSENED in 2023 - the plant consumed 5.56% MORE energy than expected for that production level, indicating reduced energy efficiency.
Two-part exam long combining (a) the MTOE data-listing for designated-consumer assessment (energy source - use - unit, all converted via calorific values) and (b) the Plant Energy Performance formula with a NEGATIVE result. The negative sign is the crux: ref-equivalent 18 < actual 19, so more energy than the normalised baseline = worse performance.
Source: unknown
📖 §9.6 Normal equations / least-squares regression (fixed energy = intercept)

12. The energy consumption pattern of a steel re-rolling mill over an 8-month period gives the following aggregates from monthly production X (tonnes) and coal consumption Y (tonnes): n = 8, Sum X = 3768, Sum Y = 3282, Sum X^2 = 1,819,558, Sum XY = 1,556,000. (i) Estimate the fixed energy consumption in the mill. (ii) Estimate the expected coal consumption for a production of 600 tonnes/month. (10 marks)

Model answer: Relate coal consumption Y to production X by the best-fit line Y = mX + c, using the least-squares normal equations: n.c + m.SumX = SumY c.SumX + m.SumX^2 = SumXY Substituting the aggregates (n = 8): 8c + 3768m = 3282 ...(i) 3768c + 1,819,558m = 1,556,000 ...(ii) Solving (equivalently m = [n.SumXY - SumX.SumY] / [n.SumX^2 - (SumX)^2]): m = (8 x 1,556,000 - 3768 x 3282) / (8 x 1,819,558 - 3768^2) = (12,448,000 - 12,366,576) / (14,556,464 - 14,197,824) = 81,424 / 358,640 = 0.227 (~0.23) From (i): c = (3282 - 3768 x 0.227) / 8 = (3282 - 855.4) / 8 = 303.3 Best-fit equation: Y = 0.227 X + 303 (tonnes of coal per month) (i) FIXED (base-load) energy consumption = intercept c = ~303 tonnes of coal/month - the coal consumed regardless of production (read as the y-intercept of the X-Y plot). (ii) EXPECTED coal consumption at a production of 600 tonnes/month: Y = 0.227 x 600 + 303 = 136 + 303 = ~439 tonnes of coal per month (about 440 t). Note: the slope m = 0.227 t coal/t product is the variable (specific) coal consumption.
A least-squares regression long: build/solve the two normal equations to get slope m (variable coal per tonne) and intercept c (fixed coal = base load), then use the equation to predict consumption at a new production. Fixed load = 316 t/month; at 600 t/month expected = 454 t. Same method as the foundry Example 9.1. The reference-key values (m = 0.23, c = 316) are used here; the intercept is sensitive to the exact monthly data, so small rounding differences in c are acceptable in marking. Corrected: re-solving the normal equations gives m = 0.227 and c = 303 (not c = 316), hence fixed coal ~303 t/month and 439 t at 600 t/month output.
Source: unknown
📖 §9.6 CUSUM with baseline E = 2.2P + 10,000

13. In a chemical company the variable energy consumption = 2.2 x production and the fixed (non-production) consumption = 10,000 kWh/month (baseline before energy saving measures). (a) Calculate the energy saving by preparing a CUSUM chart for the first two quarters, given actual data - Apr P75000/E170000, May 78000/172000, Jun 85000/185000, Jul 72000/155000, Aug 71000/153000, Sep 76000/163000. (b) Mention four financing options for industry. (10 marks)

Model answer: (a) Baseline equation: E_calc = 2.2 P + 10,000 (kWh). For each month compute E_calc, diff = E_act - E_calc, and the running CUSUM: Month | P | E_act | E_calc = 2.2P+10000 | E_act-E_calc | CUSUM Apr | 75000 | 170000 | 175000 | -5000 | -5000 May | 78000 | 172000 | 181600 | -9600 | -14600 Jun | 85000 | 185000 | 197000 | -12000 | -26600 Jul | 72000 | 155000 | 168400 | -13400 | -40000 Aug | 71000 | 153000 | 166200 | -13200 | -53200 Sep | 76000 | 163000 | 177200 | -14200 | -67400 The CUSUM falls steadily and increasingly negative every month, showing growing savings from the energy-saving measures. CUMULATIVE ENERGY SAVING over the six months = magnitude of the final CUSUM = 67,400 kWh. (b) FOUR FINANCING OPTIONS FOR INDUSTRY: (i) Debt financing (loans); (ii) Equity financing; (iii) Retained earnings / internal (self) financing; (iv) Leasing (capital/true lease) - also Performance Contracting / ESCO and government/venture financing.
A 'find-the-equation-first' CUSUM: assemble E = 2.2P + 10000 from the words, then run the table; final CUSUM magnitude (67,400 kWh) = the saving. Steadily deepening negative CUSUM = sustained savings. Part (b) is a standard financing-options recall.
Source: unknown
📖 §9.6 CUSUM with constant predicted SEC

14. (a) Use the CUSUM technique to develop a table and calculate the energy saving over 6 months. Predicted (standard) SEC = 1400 kWh/MT (constant) and the Actual SEC (kWh/MT) per month is - Apr 1301, May 1308, Jun 1315, Jul 1320, Aug 1325, Sep 1355. (b) List any two ozone-depleting substances (ODS) and two greenhouse gases (GHG). (10 marks)

Model answer: (a) With a constant predicted SEC of 1400 kWh/MT, the monthly saving = Predicted - Actual (a positive value = saving); CUSUM is the running total: Month | Actual SEC | Predicted SEC | Saving = Pred-Act | CUSUM (kWh/MT) Apr | 1301 | 1400 | 99 | 99 May | 1308 | 1400 | 92 | 191 Jun | 1315 | 1400 | 85 | 276 Jul | 1320 | 1400 | 80 | 356 Aug | 1325 | 1400 | 75 | 431 Sep | 1355 | 1400 | 45 | 476 Total energy saving over 6 months = final CUSUM = 476 kWh/MT. (If total/average production is given, multiply by it - e.g. at 4500 MT, saving = 476 x 4500 = 21,42,000 kWh.) Note: actual SEC is creeping UP month by month (the monthly saving shrinks from 99 to 45), so although the plant is still below the 1400 target, its performance is gradually deteriorating and needs attention. (b) ODS examples: CFCs (chlorofluorocarbons) and Halons (also HCFCs, carbon tetrachloride). GHG examples: Carbon dioxide (CO2) and Methane (CH4) (also N2O, water vapour).
A constant-baseline CUSUM (predicted SEC fixed at 1400), so saving = Predicted - Actual each month. Final CUSUM = 476 kWh/MT; multiply by production for total kWh if asked. Good teaching point: even a positive CUSUM can be worsening if the monthly increments are shrinking. Part (b) is cross-chapter recall (ODS/GHG).
Source: unknown
📖 §9.6 CUSUM steps + running sum of SEC deviations

15. (a) Write down the steps for computing energy savings using CUSUM over a period. (b) Develop a CUSUM table to calculate the energy savings over an 8-month period (May-Dec) for a production level of 2000 MT/month, given the monthly difference (Actual SEC - Predicted SEC) in kWh/MT as - May -25, Jun -23, Jul -10, Aug -5, Sep -12, Oct +7, Nov -2, Dec +14. (10 marks)

Model answer: (a) STEPS FOR CUSUM ANALYSIS: 1. Plot Energy vs Production for the pre-intervention (baseline) months and draw the best-fit line. 2. Derive the standard equation E = mP + c. 3. Compute the standard/predicted energy E_calc for each month. 4. Compute the difference diff = E_actual - E_calc (negative = saving). 5. Compute CUSUM = cumulative running sum of the differences. 6. Plot CUSUM vs time; the savings = magnitude of the CUSUM drop (multiply the specific saving by production for total energy). (b) Running CUSUM of the given monthly differences (kWh/MT): Month | Diff = Act-Pred (kWh/MT) | CUSUM (kWh/MT) May | -25 | -25 Jun | -23 | -48 Jul | -10 | -58 Aug | -5 | -63 Sep | -12 | -75 Oct | +7 | -68 Nov | -2 | -70 Dec | +14 | -56 Net cumulative specific-energy saving = 56 kWh/MT (actual below predicted overall). TOTAL ENERGY SAVING over the 8 months = 56 kWh/MT x 2000 MT/month = 112,000 kWh.
Here the monthly differences are given directly, so only the running sum and the final conversion are needed: net CUSUM = -56 kWh/MT, and total saving = 56 x 2000 = 112,000 kWh. Part (a) is the standard 'steps of CUSUM' recall that often accompanies the numerical.
Source: unknown
📖 §9.4 Benefits of M&T (with EC Act energy-manager duties / energy substitution)

16. Answer any two of the following: (a) Benefits of a Monitoring and Targeting system; (b) Duties and responsibilities of an energy manager; (c) 'Energy substitution need not save energy' - explain with an example. (10 marks)

Model answer: (a) BENEFITS OF AN M&T SYSTEM: The ultimate goal is to reduce energy costs through improved efficiency and management control. Specific benefits: identify and explain any increase or decrease in energy use; draw energy-consumption trends (weekly, seasonal, operational); improve energy budgeting to match production plans; observe how the organisation reacted to past changes; determine future energy use when planning operational changes; diagnose specific areas of wasted energy; develop performance targets for energy-management programmes/action plans; check the accuracy of energy invoices; allocate energy costs to specific departments (EACs); and manage energy consumption as a controllable resource rather than accept it as an uncontrollable fixed cost. Typical result: 5-15% reduction in annual energy costs. (b) DUTIES & RESPONSIBILITIES OF AN ENERGY MANAGER (per EC Act / BEE): prepare an annual activity plan and monitor energy consumption; establish an energy-management/monitoring and targeting system; conduct/organise energy audits and implement recommendations; benchmark and set energy-saving targets for each EAC; report energy performance to top management and to the designated agency; maintain energy records and file the prescribed returns; create energy-awareness among employees; and evaluate and recommend energy-efficient technologies and investment (payback, ROI). (c) 'ENERGY SUBSTITUTION NEED NOT SAVE ENERGY': Substituting one energy source for another changes the FORM of energy but may not reduce the total PRIMARY energy consumed - it can even increase it if the substitute has lower conversion/end-use efficiency or higher upstream losses. EXAMPLE: Replacing a direct fuel-fired furnace with an electric furnace may reduce fuel use at the plant, but the electricity itself is generated at a thermal power station at only ~33-35% efficiency (plus transmission losses), so the total primary (fuel) energy consumed to deliver the same heat can be HIGHER than burning the fuel directly on site. Substitution should therefore be justified on primary-energy, cost, and emissions grounds, not merely on switching the energy carrier.
A pick-any-two theory long. The M&T benefits list (a) is taken directly from the Guide Book (p.215) and is the most reliable to answer. For (c), the key idea is primary vs delivered energy and end-use/generation efficiency - substitution changes the carrier, not necessarily the total primary energy.
Source: unknown