- Email: [email protected]
Multi-criteria evaluation of cooking devices with special reference to utility of parabolic solar cooker (PSC) in India
Energy 31 (2006) 1215–1227 www.elsevier.com/locate/energy
Multi-criteria evaluation of cooking devices with special reference to utility of parabolic solar cooker (PSC) in India S.D. Pohekara,*, M. Ramachandranb a
Center for Renewable Energy and Environment Development (CREED), Birla Institute of Technology and Science (BITS), Pilani 333 031, India b BITS, Pilani-Dubai Campus, P.O. Box 500022, Block No 11, Knowledge Village, Dubai, UAE Received 18 February 2004
Abstract Multi-criteria decision making is an emerging technique for evaluation and policy formulation for renewable energy technology promotion. In this paper, the case of the parabolic solar cooker (PSC), which is a relatively recent innovation, is evaluated with respect to eight prevalent domestic cooking devices in India. Thirty different criteria categorized under technical, economic, environmental, social, behavioral and commercial aspects are considered for the evaluation based on the additive Multi Attribute Utility Theory (MAUT) model. A survey of various decision making groups and user preferences for domestic cooking devices in India is used to formulate the evaluation matrix. Expert opinion is collected to devise the utility functions. On the basis of user preferences and expert opinion, it is found that the liquefied petroleum gas (LPG) stove has the highest utility, followed by the kerosene stove, solar box cooker (SBC) and PSC, respectively. Sensitivity analyses are carried out to identify the areas of improvement for the widespread use of PSC. q 2005 Elsevier Ltd. All rights reserved. Keywords: Parabolic solar cookers; Utility assessment; Scaling constants; Sensitivity analyses
1. Introduction In India, the domestic sector accounts for 45% of energy consumption at the national level, of which 90% is for the cooking alone [1]. Also, like other developing countries, non-commercial sources such as biomass, agricultural waste and fuel wood account for a sizeable portion of energy consumption for * Corresponding author. Tel.: C91 1596 245783; fax: C91 1596 244183. E-mail address: [email protected] (S.D. Pohekar). 0360-5442/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.energy.2005.04.012
1216
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
cooking needs. The consumption is increasing due to population growth. This has led to environmental problems associated with deforestation and fuel-wood shortage. Solar cookers and their variations, such as, PSC are low cost options for meeting the cooking energy needs as well as environmental protection. The PSC is an emerging device which has a great potential in India. However, PSC technology will have to compete with prevalent cooking devices in the country. Even though over 5,00,000 SBCs have been sold so far, the large potential of solar cookers is yet untapped [2]. Limited outreach of the technology is attributed to technical limitations, costs, awareness, several behavioral issues, distribution and service network, etc. The Ministry of Non-conventional Energy Sources (MNES), Government of India has devised innovative schemes for promoting the use of PSC. Success of such schemes will be determined by future acceptance of these technologies by the users. Utility assessment is likely to give feedback to the planners and policy makers for formulating strategies for further commercialization. In this paper, the utility of PSC in India with respect to eight prevalent cooking devices has been computed by knowing users’ preferences and expert opinion on thirty different criteria and additive MAUT framework.
2. Prevailing evaluation practices The history of solar cookers goes back a long time. The first reference to solar cooking was that of Nicholas-de-Saussure (1740–1799). There are 50 major designs of solar cookers and more than 100 variations, of which many are already patented [3]. The evaluation of solar cookers has been attempted by various researchers since 1960. In 1961, Volunteers for Technical Assistance (VITA) attempted a detailed review of solar cookers considering such criteria as cooking performance, durability, cost, weight, portability, ease of operation, ease of manufacture and adoptability to local skills and materials [4]. Technical evaluation of various solar cooker designs is attempted by various authors [5,6]. Some of the studies attempted to assess the marketability of SBCs. Amongst these, the review by Gujarat Energy Development Agency (GEDA) [7] is important. Bowman and Blatt [8] attempted the assessment of various solar cookers considering various techno-economic-psychological parameters on 100 point scale. Sharan and Naik [9] attempted to discover the socio-psychological factors determining the acceptance of SBCs in India. The PSC has been introduced in Indian climatic conditions in 1999. The technical evaluation conducted at Technical Backup Units (TBUs) for solar thermal devices for MNES [10] have reported their overall performance to be satisfactory. The present evaluation attempts to quantify the comprehensive utility of PSC vis-a`-vis other cooking devices. To this effect, various criteria and devices are identified. The prevailing cooking devices are assessed with these criteria using the additive MAUT framework.
3. Selection of criteria Cooking energy substitution by renewables is a burning issue and is discussed widely in the literature. Acceptance studies on solar cookers in various parts of India, diagnostic case studies for solar cookers, conjoint analysis for solar cookers and experiences of energy substitution in other developing economies
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1217
are studied to understand end users feedback and simultaneously identify the criteria for the present evaluation. Table 1 presents a summary of 30 criteria and their significance in evaluation of a cooking device in general and PSC in particular discussed in the above studies. Major aspects of assessment are classified as technical, economic, social/environmental, behavioral, and commercial. Fig. 1 presents twenty one criteria of qualitative and nine of quantitative nature. Fuel consumption, cooking time, durability, size/weight/space needs and various costs involved are identified to be quantitative in nature.
4. Selection of devices A large variety of cooking devices are used in India based on prevailing climatic conditions and the socio-economic settings. Traditional low cost devices like chulhas (cook-stoves) are widely used by rural masses which constitute about 68% of the total population. These devices have very low thermal efficiencies (10–15%) and are highly polluting. Also, there are operational difficulties faced by rural women. With the increasing scarcity of fuel-wood, rural women have to walk up to 10 km and spend 3–4 h a day for its collection. Improved chulhas have durability of 2–4 years and require fuel-wood in lesser quantity. However, of 23 Million improved chulhas installed in India, only 6 Million are found to be functional [20]. Biogas plants need higher initial investment and trained manpower for installation. Many biogas plants are non-functional due to non-availability of water throughout the year. The problems identified for limited use of improved chulhas and biogas stoves are operational, social and behavioral leading to non-participation of masses. Kerosene stoves have good thermal efficiency, benefits of simplicity and availability due to good market network and hence are used in semi-urban areas. Due to high costs and weak supply chain in rural India, only 1.3% of rural houses use LPG compared with 27.2% in urban homes [21]. There is pressure to increase the subsidy of kerosene and LPG to further reduce the prices. The subsidies are posing a heavy burden on overall economy of the country. Modern devices such as micro-wave ovens and electric ovens are not affordable by the masses due to high capital costs, high operating costs and intermittent electrical supply. In view of the above problems, solar cooking can be a sustainable energy option for cooking needs. However, in the foreseeable future, PSC technology will have to compete with conventional cooking energy technologies. The PSC considered for evaluation in the present review can deliver 0.6 kW of thermal power with a thermal efficiency of 55–60% and stagnation temperature of 350–400 8C [10]. Thus the cooker is useful for boiling, roasting, frying and baking operations. The cooker is user friendly, convenient and can meet the needs of around 15 people. Community versions of PSC with 7–9 m2 of reflector area and an average efficiency of 25–30% are also installed in India.
5. Multi-criteria evaluation Since majority of the criteria are qualitative in nature, the evaluation on these criteria is conducted through a survey. Since there are 30 criteria involved in the problem, two questionnaires in the form of tabular columns are found to be suitable; one for recording the importance of criteria and the other for recording the performance of identified devices. The questionnaires are tested, validated, modified and opinion is collected from thirty experts. Since there are a wide range issues, from technical
1218
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
Table 1 Significant criterion in cooking energy decisions Criteria no and name
Desired value
Significance
References
CR 1. Fuel consumption CR 2. Cooking timea CR 3. Durabilitya CR 4. Quality, reliability CR 5. Sophistication level CR 6. Size/Weight/Space needsa CR 7. Ruggedness CR 8. Continuity of use CR 9. Need for tracking CR 10. Nutrition value of food CR 11. Initial costa
Low Low High High High Low
Reduced cost, Reduced fuel collection time Higher efficiency, More convenience, Lower nutrient value Reduced costs, Higher reliability, Lower maintenance Lower maintenance Better control of heat, convenience Increased suitability, lower bulk of the device
[1,11] [12,13] [9,11] [9,12] [14,15] [9,11]
High High Low High
Lower maintenance Reduced need of additional system, More convenience Only for solar cookers, Efficiency, Reduced cooking time Healthy food
[9,11] [11,17] [11,16] [13]
Low
[11,13]
CR 12. Fuel cost per montha
Low
CR 13. Maintenance costa
Low
CR 14. Available subsidya CR 15. Rate of interest on loan if anya CR 16. Pollution hazards CR 17. Human drudgery
Low Low
Easy commercialization, Increased Reliability, Convenience Easy commercialization, Increased Reliability, Convenience Easy commercialization, Increased Reliability, Convenience Reduced quality, Easy commercialization Easy commercialization
[1,17,19] [1,19]
High High High High High High High Low
Increased social utility Convenience, Increased productive time for earning for family, education etc Increased social utility Increased commercialization, Higher motivation to use Increased commercialization Increased suitability Versatile cooking, Convenience, Higher motivation to use Versatile cooking, Convenience, Higher motivation to use Versatile cooking, Convenience, Higher motivation to use Convenience, Reduced Costs, Increased Reliability
High
Commercialization, Convenience, Higher motivation to use
[15,19]
High
Increased reliability, Commercialization
[15,18]
High High Low
Better commercialization Better commercialization Convenience, Higher motivation to use
[18] [18] [15,16]
a
CR 18. Overall safety CR 19. Aesthetics CR 20. Motivation to buy CR 21. Taste of food CR 22. Cleanliness of utensils CR 23. Ease of operation CR 24. Type of dishes cooked CR 25. Need for additional cooking system CR 26. Improvement in models CR 27. Spares and after sales service CR 28. Distribution network CR 29. Market research CR 30. Need for user training a
Low Low
[11,13] [11,13] [11,13] [15,18]
[1,17,19] [9,18] [18] [11,14] [11,14] [11,14] [11,14] [14,15]
Quantitative criteria.
improvements to need for user training, experts are identified from various groups. The judgments of energy educators as awareness creators (10), policy makers (10), and researchers as technology up graders (10) are collected though postal, e-mail and personal interactions. The preferences of actual users of the identified devices (30) also are considered in the study. Assessment of scaling constants is
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1219
Fig. 1. Classification of criteria.
determined by collecting the judgments of thirty experts on a 10 point qualitative linear scale (10 representing very high importance and 2 representing very low importance). Weighted average scores are computed for importance of criteria by using a custom built interface developed in a Visual Basic environment. The scaling constants indicate weighted average importance of criteria and are calculated as follows. Pj Scaling$constant Z
i
Wj Nj N
(1)
1220
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
Where Wj indicates the importance of jth criteria (on assigned scale) by the numbers of respondents Nj for that criteria and N are total numbers of responses. The judgments of experts and users are found to be consistent with each other and are indicated in last column of Table 2. The present analyses use the aggregate opinion of all the respondents. The evaluation matrix (Table 2) also is formulated by collecting opinions from experts, users and the published literature. Qualitative criteria judgments are aggregate based on opinion of all the respondents. The values for quantitative criteria are directly taken from published literature [1,10, 21–23]. The values of 2, 4, 6, 8 and 10 are allotted to indicate values of performance of a device as very low, low, moderate, high and very high, respectively. For example, need of user training is very
Table 2 Evaluation matrix for all criteria Device no criteria
Chulha (1)
Improved chulha (2)
Kerosene stove (3)
Biogas stove (4)
LPG stove (5)
Micro wave oven (6)
Electric oven (7)
Solar box cooker (8)
Parabolic solar cooker (9)
Scaling constant
CR 1.a CR 2.a CR 3.a CR 4. CR 5. CR 6.a CR 7. CR 8. CR 9. CR 10. CR 11.a CR 12.a CR 13.a CR 14.a CR 15.a CR 16. CR 17. CR 18. CR 19. CR 20. CR 21. CR 22. CR 23. CR 24. CR 25. CR 26. CR 27. CR 28. CR 29. CR 30.
2 60 1 2 2 2 2 10 0 6 10 20 0 0 0 10 10 2 2 2 6 2 10 10 2 2 10 10 2 2
1 60 4 2 2 1 2 10 0 6 50 10 0 50 0 10 10 2 2 2 6 2 10 10 2 2 10 6 2 2
0.5 30 15 8 6 2 6 10 0 6 200 100 50 0 0 8 6 4 4 6 2 2 10 10 4 6 10 6 8 4
0 15 5 4 6 50 6 4 0 6 5000 0 200 2000 0 2 6 10 4 6 6 8 4 10 4 4 4 6 2 8
0.25 15 20 10 10 10 10 10 0 6 4000 250 50 0 13 4 2 6 10 10 6 10 8 10 6 10 10 2 10 4
2 5 5 10 10 5 8 10 0 2 8000 200 200 0 13 10 2 2 10 8 2 10 4 2 8 10 6 6 10 10
2 30 5 6 10 3 8 10 0 2 5000 400 200 0 13 10 2 2 8 4 10 10 4 2 10 6 4 2 4 6
0 180 10 6 4 5 8 4 4 10 2000 0 50 500 3 0 2 10 8 4 10 10 6 2 10 6 4 2 4 6
0 20 20 6 4 15 8 2 10 8 7000 0 20 2000 3 0 2 8 8 4 10 10 4 4 10 6 2 2 4 8
0.0355 0.0384 0.0397 0.0376 0.0326 0.0332 0.0310 0.0335 0.0326 0.0371 0.0371 0.0293 0.0274 0.0252 0.0230 0.0217 0.0296 0.0433 0.0348 0.0314 0.0381 0.0368 0.0387 0.0332 0.0306 0.0342 0.0352 0.0355 0.0315 0.0323
a
Quantitative criteria.
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1221
low for chulha and hence assigned 10 points, whereas it is very high in case of microwave ovens which is assigned 2 points. Since biogas stove, SBC and PSC are dependent on renewable energy, fuel consumption and fuel costs are taken as zero. On the other hand, need for tracking is assigned zero value in the evaluation matrix for all other devices except SBC and PSC, as the energy input for cooking with these devices is independent of the Sun’s position. Many of the cooking devices require lower capital investments and are not given any subsidy. The relevant score assigned is zero for these devices. The quantitative scores in evaluation matrix indicate quantity of fuel consumed per family per day in kilograms, cooking time in minutes, durability in years and various costs in Rupees. Size/weight /space needs are indicated by the overall bulk in kilograms whereas rate of interest on loan is indicated in percent (Table 3). Table 3 Normalized evaluation matrix Device no criteria
Chulha (1)
Improved chulha (2)
Kerosene stove (3)
Biogas stove (4)
LPG stove (5)
Micro wave oven (6)
Electric oven (7)
Solar box cooker (8)
Parabolic solar cooker (9)
CR 1. CR 2. CR 3. CR 4. CR 5. CR 6. CR 7. CR 8. CR 9. CR 10. CR 11. CR 12. CR 13. CR 14. CR 15. CR 16. CR 17. CR 18. CR 19. CR 20. CR 21. CR 22. CR 23. CR 24. CR 25. CR 26. CR 27. CR 28. CR 29. CR 30.
0.0000 0.6857 0.0000 0.0000 0.0000 0.9796 0.0000 1.0000 1.0000 0.5000 1.0000 0.9500 1.0000 1.0000 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.5000 0.0000 1.0000 1.0000 1.0000 0.0000 1.0000 1.0000 0.0000 1.0000
0.5000 0.6857 0.1579 0.0000 0.0000 1.0000 0.0000 1.0000 1.0000 0.5000 0.9950 0.9750 1.0000 0.9800 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.5000 0.0000 1.0000 1.0000 1.0000 0.0000 1.0000 0.5000 0.0000 1.0000
0.7500 0.8571 0.7368 0.7500 0.5000 0.9796 0.5000 1.0000 1.0000 0.5000 0.9762 0.7500 0.7500 1.0000 1.0000 0.2000 0.5000 0.2500 0.2500 0.5000 0.0000 0.0000 1.0000 1.0000 0.7500 0.5000 1.0000 0.5000 0.7500 0.7500
1.0000 0.9429 0.2105 0.2500 0.5000 0.0000 0.5000 0.2500 1.0000 0.5000 0.3755 1.0000 0.0000 1.0000 0.7692 0.8000 0.5000 1.0000 0.2500 0.5000 0.5000 0.7500 0.0000 1.0000 0.7500 0.2500 0.2500 0.5000 0.0000 0.2500
0.8500 0.9429 1.0000 1.0000 1.0000 0.8163 1.0000 1.0000 1.0000 0.5000 0.5006 0.3750 0.7500 0.0000 0.0000 0.6000 1.0000 0.5000 1.0000 1.0000 0.5000 1.0000 0.7500 1.0000 0.5000 1.0000 1.0000 0.0000 1.0000 0.7500
0.0000 1.0000 0.2105 1.0000 1.0000 0.9184 0.7500 1.0000 1.0000 0.0000 0.0000 0.5000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 1.0000 0.7500 0.0000 1.0000 0.0000 0.0000 0.2500 1.0000 0.5000 0.5000 1.0000 0.0000
0.0000 0.8571 0.2105 0.5000 1.0000 0.9592 0.7500 1.0000 1.0000 0.0000 0.3755 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 0.7500 0.2500 1.0000 1.0000 0.0000 0.0000 0.0000 0.5000 0.2500 0.0000 0.2500 0.5000
1.0000 0.0000 0.4737 0.5000 0.2500 0.9184 0.7500 0.2500 0.6000 1.0000 0.7509 1.0000 0.7500 0.7500 0.7692 1.0000 1.0000 1.0000 0.7500 0.2500 1.0000 1.0000 0.3300 0.0000 0.0000 0.5000 0.2500 0.0000 0.2500 0.5000
1.0000 0.9143 1.0000 0.5000 0.2500 0.7143 0.7500 0.0000 0.0000 0.7500 0.1252 1.0000 0.9000 1.0000 0.7692 1.0000 1.0000 0.7500 0.7500 0.2500 1.0000 1.0000 0.0000 0.2500 0.0000 0.5000 0.0000 0.0000 0.2500 0.7500
1222
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
6. The additive MAUT model There are various methods to assess the preferences, such as Analytic Hierarchy Process (AHP), Simple Multi-Attribute Rating Technique (SMART), etc. While SMART is a simple method for assessing preferences, it has limitations in handling inconsistent units. The AHP model involves higher complexity with 30 criteria and consistency checks at each level. The advantage of the additive MAUT is its simplicity. It combines the main advantages of simple scoring techniques and optimization models. Utility is a measure of desirability or satisfaction and provides a uniform scale to compare and/or combine tangible and intangible criteria. The MAUT theory is developed to help decision makers assign utility values to devices in terms of single attribute utility functions and combine individual evaluations to obtain overall utility values. It considers the decision maker’s preferences in the form of a utility function which is defined over a set of attributes. However, it is observed that MAUT is not very extensively used in energy planning [24]. This may be due to requirements of interactive decision environment required for formulating utility functions and complexity of computing scaling constants using the MAUT algorithm. Selecting portfolios for solar energy projects [25] and energy policy making [26] are the applications identified in the literature. The utility value can be determined by the following three steps: (i) Determination of single attribute utility functions. The values of utilities vary between zero and one and reflect the level of importance in the achievement of that attribute. (ii) Verification of preferential and utility independence conditions. (iii) Derivations of the multi attribute utility function to determine intermediate utilities. All decisions involve choosing one or a few, from several, alternatives. Each alternative is assessed for desirability on a number of scored criteria. Utility function connects the criteria scores with desirability. The utility functions are of two important forms, viz. additive and multiplicative. The most common formulation of a multi-criteria utility function is the additive model. Ui O Wj ; uij ;
for all i;
(2)
where Ui is the overall utility value of device i uij is the utility value of the jth criterion for the ith device Uij equals u (Xi), for 1RiRn and iRjRm Xi equals (xij) for 1RiRn and 1RjRm Xi designates a specific value of xij n is total number of criteria, m is total number of devices Wj is the scaling constant of jth criteria
7. Results and discussions An analysis of scaling constants indicate that utility of the selected cooking energy devices is governed by technical (0.352) followed by behavioral (0.246), commercial (0.167) and economic criteria (0.141). The sum of all scaling constants is unity. It also is observed that social/environmental
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1223
criteria are assigned lower importance (0.095), overall safety (0.0433), durability (0.0397) and ease of operation (0.0387) also received low importance ranking by the respondents. Pollution hazard (0.0217) is ranked the least important. An analysis of the evaluation matrix reveals the strengths of PSC as fuel consumption, durability, nutritional value of food, fuel cost, available subsidy, pollution hazards, human drudgery, taste of food, cleanliness of utensils etc. The SBCs are also indicated to have similar strengths, except cooking time. The LPG stove has higher values for all the criterion except initial, fuel cost, rate of interest on loan, distribution network. Values for chulha and improved chulha indicate strengths in terms of continuity of use, ease of operation, type of dishes cooked, spares and after sales service. Kerosene stoves have moderate values for most of the criteria. Micro-wave and electric ovens have strengths for many criteria except costs and need for user training. Following the formulation of the evaluation matrix, the next step is identification of the best and worst outcomes (criteria scores) for each of the criteria. The best outcome for a particular criterion is assigned utility value of unity and the worst as zero. To assign intermediate values, utility functions are developed by the technique known as standard gambling [27]. The decision maker is offered certain outcome with probability pZ1. Risk option with a probabilistic outcome in the form of gamble with best outcome considering probability p or worst outcome with probability of 1-p is also offered to the experts. Fig. 2 shows the pair of lotteries for criterion ‘motivation to buy (CR 20)’. Lottery 1 and Lottery 2 shows best consequences of the same criterion with certain probability of p and worst probability of 1-p respectively. The decision maker in the present case selected the indifference probability of pZ0.5 after mental trial and error. This indifference allows assessing the utility values from the principle of expected utility from the probability theory. For all the criteria, linear utility functions are considered to assess the utilities. There are two kinds of criteria in the present problem viz; the maximization criteria (the maximum value is desirable e g. durability) and the minimization criteria (the minimization value is desirable, e g. initial cost). Since the present problem uses probability of 0.5 and linear utility functions, the intermediate utility values are thus obtained by normalizing the evaluation matrix as follows.
uðjÞ Z
Aj K Amin Amax K Amin
For maximization criteria
Fig. 2. Pair of Lotteries for motivation to buy (CR 20).
(3)
1224
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
uðjÞ Z
Amax K Aj Amax K Amin
For minimization criteria
(4)
Where Aj represents the score assigned to cooking device in evaluation matrix. Amax and Amin are the maximum and minimum scores assigned for the selected criteria for identified devices. Following the above normalization of evaluation matrix is computed. The final utilities are computed by multiplying the normalized utility values (on the scale of zero to one) by respective scaling constants. The final utility Table 4 Utility values for identified devices Device no criteria
Chulha (1)
Improved chulha (2)
Kerosene stove (3)
Biogas stove (4)
LPG stove (5)
Micro wave oven (6)
Electric oven (7)
Solar box cooker (8)
Parabolic solar cooker (9)
CR 1. CR 2. CR 3. CR 4. CR 5. CR 6. CR 7. CR 8. CR 9. CR 10. CR 11. CR 12. CR 13. CR 14. CR 15. CR 16. CR 17. CR 18. CR 19. CR 20. CR 21. CR 22. CR 23. CR 24. CR 25. CR 26. CR 27. CR 28. CR 29. CR 30. Additive utility Ranking
0.0000 0.0263 0.0000 0.0000 0.0000 0.0325 0.0000 0.0335 0.0326 0.0186 0.0371 0.0278 0.0274 0.0252 0.0230 0.0000 0.0000 0.0000 0.0000 0.0000 0.0191 0.0000 0.0387 0.0332 0.0306 0.0000 0.0352 0.0355 0.0000 0.0323 0.5086
0.0178 0.0263 0.0063 0.0000 0.0000 0.0332 0.0000 0.0335 0.0326 0.0186 0.0369 0.0286 0.0274 0.0247 0.0230 0.0000 0.0000 0.0000 0.0000 0.0000 0.0191 0.0000 0.0387 0.0332 0.0306 0.0000 0.0352 0.0178 0.0000 0.0323 0.5156
0.0266 0.0329 0.0293 0.0282 0.0163 0.0325 0.0155 0.0335 0.0326 0.0186 0.0362 0.0220 0.0206 0.0252 0.0230 0.0043 0.0148 0.0108 0.0087 0.0157 0.0000 0.0000 0.0387 0.0332 0.0230 0.0171 0.0352 0.0178 0.0236 0.0242 0.6600
0.0355 0.0362 0.0084 0.0094 0.0163 0.0000 0.0155 0.0084 0.0326 0.0186 0.0139 0.0293 0.0000 0.0252 0.0177 0.0174 0.0148 0.0433 0.0087 0.0157 0.0191 0.0276 0.0000 0.0332 0.0230 0.0086 0.0088 0.0178 0.0000 0.0081 0.5127
0.0302 0.0362 0.0397 0.0376 0.0326 0.0271 0.0310 0.0335 0.0326 0.0186 0.0186 0.0110 0.0206 0.0000 0.0000 0.0130 0.0296 0.0217 0.0348 0.0314 0.0191 0.0368 0.0290 0.0332 0.0153 0.0342 0.0352 0.0000 0.0315 0.0242 0.7581
0.0000 0.0384 0.0084 0.0376 0.0326 0.0305 0.0233 0.0335 0.0326 0.0000 0.0000 0.0147 0.0000 0.0000 0.0000 0.0000 0.0296 0.0000 0.0348 0.0236 0.0000 0.0368 0.0000 0.0000 0.0077 0.0342 0.0176 0.0178 0.0315 0.0000 0.4849
0.0000 0.0329 0.0084 0.0188 0.0326 0.0318 0.0233 0.0335 0.0326 0.0000 0.0139 0.0000 0.0000 0.0000 0.0000 0.0000 0.0296 0.0000 0.0261 0.0079 0.0381 0.0368 0.0000 0.0000 0.0000 0.0171 0.0088 0.0000 0.0079 0.0162 0.4162
0.0355 0.0000 0.0188 0.0188 0.0082 0.0305 0.0233 0.0084 0.0196 0.0371 0.0279 0.0293 0.0206 0.0189 0.0177 0.0217 0.0296 0.0433 0.0261 0.0079 0.0381 0.0368 0.0128 0.0000 0.0000 0.0171 0.0088 0.0000 0.0079 0.0162 0.5805
0.0355 0.0351 0.0397 0.0188 0.0082 0.0237 0.0233 0.0000 0.0000 0.0278 0.0046 0.0293 0.0247 0.0252 0.0177 0.0217 0.0296 0.0325 0.0261 0.0079 0.0381 0.0368 0.0000 0.0083 0.0000 0.0171 0.0000 0.0000 0.0079 0.0242 0.5637
7
5
2
6
1
8
9
3
4
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1225
value is found by adding the utilities of devices on different criteria as follows. Utility Z
30 X
Wj uij
(5)
1
Table 4 shows utility values and ranking for identified devices following this methodology. The utility of a LPG stove is found to be the highest at 0.7581 followed by kerosene stove (0.6600), SBC (0.5805) and PSC (0.5637) on a scale between 0 and 1. An electric stove has the least utility of 0.4162. The LPG stove is a preferred device because of its strengths on most parameters. The strengths of PSC and SBC are on environmental and social issues for which the scaling constants are the smallest. Sensitivity analyses are conducted to formulate strategies for increasing the utility of PSC (Table 5). The utility is sensitive to technical, as well as behavioral issues. Alteration of economic criteria does not have much effect on the utility. The problem of dissemination appears to be sensitive to behavioral issues such as type of dishes cooked and convenience. Certain commercial parameters such as availability of spares and after sales service and improvement in existing models are also found to be key factors in increasing the overall utility of PSC in the Indian domestic sector. The potential strategies for better dissemination may be similar for PSC and SBC. Reduction in size/weight/space improves the utility value. Making PSC a trendy and portable device can be achieved by using aluminium, fiber composites to mount reflectors and caster wheels for easy movement. These measures however do not make any significant change in ranking of PSC. Improvements in quality/reliability of the product can be a good alternative strategy. Better quality systems however warrant large volumes of production to be cost effective. A competing device using that strategy is the kerosene stove. Attempts have been made to check the sensitivity to a few behavioral criteria, viz. motivating potential buyers, maximizing ease of operation and type of dishes cooked. The studies indicate changes in the ranking pattern. This can be achieved by creating awareness amongst the potential users, and arranging cooking demonstrations etc. Addressing all behavioral issues simultaneously may make substantial change in the ranking pattern. Table 5 Rankings of PSC for alternative strategies No
Alternative strategy
Criteria no
Utility
Rank
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
Reducing the bulk weight, size etc Improving aesthetics Improvement in existing models Improving quality and reliability Extensive market research Reducing cost to Rs.2000 Motivating potential buyers Cooking demos for type of dishes cooked Improving after sales service Improvements in distribution network Maximizing ease of operation Improving on all the technical criteria Improving on all the commercial criteria Improving on all the behavioral criteria Improvements on all criteria
6 19 26 4 29 11 20 24 27 28 23 1–10 26–30 19–25 1–30
0.5427 0.5724 0.5808 0.5825 0.5837 0.5869 0.5872 0.5886 0.5989 0.5992 0.6024 0.6642 0.6832 0.6901 1.0000
4 4 3 3 3 3 3 3 3 3 3 2 2 2 1
1226
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
Commercial criteria like improvements in aesthetics, improvements in existing models, extensive market research and good after sales service can improve the ranking pattern.
8. Conclusions There is a need to assess the utility of cooking devices by considering multiple criteria in view of prevailing techno-economic and socio-commercial scenarios. Additive MAUT provides one such framework for ranking cooking devices by considering quantitative and qualitative criteria. The utility value can be used as a yardstick for knowing perceptions of different groups involved in the decision process. The utility values obtained reveal that LPG has the highest and electric stove has the lowest utility. The SBCs occupy third position in the ladder followed by PSC. The sensitivity analyses provide better insight into the problem of lower utility of PSC in the present Indian context. The kerosene stove may emerge as a competitive option to LPG and PSC due to lower fuel needs, better cooking time, a good market and a service network supported by economics. Sensitivity analysis indicates that the problem of better dissemination of PSC is merely not an economic issue. The utility of PSC on technical, behavioral and commercial aspects should be improved. Creating awareness among the masses, building confidence based on behavioral issues, together with technical improvements in the product can be taken into account in formulating the strategies for better dissemination.
References [1] Reddy BS. Overcoming the energy efficiency gap in India’s household sector. Energy Policy 2003;31:1117–27. [2] Ministry of Non-conventional Energy Sources. Annual Report, New Delhi: Government of India. CGO Complex Lodi Road; 2003. [3] Kundapur A. Reviews of solar cooker designs. TERI Inf Digest Energy 1998;8(1):1–37. [4] Volunteers in Technical Assistance. Evaluation of Solar Cookers. VITA report. No 10. 1600 Wilson Boulevard, Suite 1030, Arlington, Virginia; 1961. [5] Das TCT, Karmarkar S, Rao DP. Solar box cooker. 1. Modeling. Solar Energy 1994;52(3):265–72. [6] Funk PA, Larson DL. Parametric model of solar cooker performance. Solar Energy 1998;62(1):63–8. [7] Gujarat Energy Development Agency. Proposal for Popularizing Solar Cookers. Vadodara: II, Floor Suraj Plaza II Sayaji Ganj; 1979. [8] Bowman TE, Blatt JH. Solar cookers, history, design, fabrication, testing and evaluation. Florida: Florida Institute of Technology; 1978. [9] Sharan Girija, Naik G. Assessing consumer preference for product features selection: solar cookers. Vikalpa 1997;22(4): 49–54. [10] Sawhney RL, Buddhi D, Gautam RP editors. Renewable energies and energy efficiency for sustainable development. Proceedings of 23rd National Renewable Energy Convention. Indore, India. Solar Energy Society of India, New Delhi; 1999. [11] Gore AP, Paranjape SA, Pandit SJ, Prayag V. Why solar cookers do not sell. Changing Villages 1990;9(4):219–25. [12] Ahmad B. Users and disusers of box solar cookers in urban India-implications for solar cooking projects. Solar Energy 2000;69(1–6 Suppl):209–15. [13] Kumar S, Kandpal TC, Mullik SC. Solar cooker use in Delhi: Pilot survey of urban households. Int J Ambient Energy 1997;18(2):77–82.
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1227
[14] Avinashilingam Institute for Home Science and Higher Education for Women. Proceedings of 3rd International Conference on Solar Cookers Use and Technology, Coimbatore, India; 1997. [15] Moulik R. Socio-psychological and economic factors affecting acceptability of solar cookers in Gujarat. Unpublished Thesis. Ahmedabad: Indian Institute of Management; 1985. [16] Sama Shrestha. Teaching and training communities of Kathmandu valley in the use of solar parabolic cookers. In: Misra Anil, editor. Renewable energy education current practices and future projections. New Delhi: Tata Energy Research Institute; 1998. p. 133–43. [17] Philip SK, Makwana HM, Singhal AK. Monitoring of subsidized solar cookers in Gujarat: a case study. SESI J 1987;1: 37–43. [18] Mohanty RC. Gujarat energy development agency: the case of solar cookers. Vikalpa 1991;16(2):65–79. [19] Biermann E, Grupp M, Palmer R. Solar cooker acceptance in South Africa: results of comparative field tests. Solar Energy 1999;66(6):401–7. [20] Neudoerfler RC, Malhotra Preeti, Ramana PV. Participatory rural energy planning in India—a policy context. Energy Policy 2001;29:371–81. [21] Tata Energy Data and Directory Yearbook (TEDDY) 2001/02. New Delhi: Tata Energy Research Institute; 2002. [22] Gupta S, Ravindranath NH. Financial analysis of cooking energy options for India. Energy Convers Manage 1997;38(10): 1869–77. [23] Ramanathan R, Ganesh LS. Multi-objective analysis of cooking-energy alternatives. Energy 1994;19(4):469–78. [24] Huang JP, Poh KL, Ang BW. Decision analysis in energy and environmental modeling. Energy 1995;20(9):843–55. [25] Golabi K, Kirkwood CW, Sicherman A. Selecting a portfolio of solar energy projects using multiattribute preference theory. Manage Sci 1981;22(2):174–89. [26] Jones M, Hope C, Hughes R. A multi-attribute value model for the study of UK energy policy. J Operations Res Soc 1990; 41(10):919–29. [27] Keeney RL, Raiffa H. Decisions with multiple objectives: preferences and value tradeoffs. Cambridge: Cambridge University Press; 1993.
Multi-criteria evaluation of cooking devices with special reference to utility of parabolic solar cooker (PSC) in India S.D. Pohekara,*, M. Ramachandranb a
Center for Renewable Energy and Environment Development (CREED), Birla Institute of Technology and Science (BITS), Pilani 333 031, India b BITS, Pilani-Dubai Campus, P.O. Box 500022, Block No 11, Knowledge Village, Dubai, UAE Received 18 February 2004
Abstract Multi-criteria decision making is an emerging technique for evaluation and policy formulation for renewable energy technology promotion. In this paper, the case of the parabolic solar cooker (PSC), which is a relatively recent innovation, is evaluated with respect to eight prevalent domestic cooking devices in India. Thirty different criteria categorized under technical, economic, environmental, social, behavioral and commercial aspects are considered for the evaluation based on the additive Multi Attribute Utility Theory (MAUT) model. A survey of various decision making groups and user preferences for domestic cooking devices in India is used to formulate the evaluation matrix. Expert opinion is collected to devise the utility functions. On the basis of user preferences and expert opinion, it is found that the liquefied petroleum gas (LPG) stove has the highest utility, followed by the kerosene stove, solar box cooker (SBC) and PSC, respectively. Sensitivity analyses are carried out to identify the areas of improvement for the widespread use of PSC. q 2005 Elsevier Ltd. All rights reserved. Keywords: Parabolic solar cookers; Utility assessment; Scaling constants; Sensitivity analyses
1. Introduction In India, the domestic sector accounts for 45% of energy consumption at the national level, of which 90% is for the cooking alone [1]. Also, like other developing countries, non-commercial sources such as biomass, agricultural waste and fuel wood account for a sizeable portion of energy consumption for * Corresponding author. Tel.: C91 1596 245783; fax: C91 1596 244183. E-mail address: [email protected] (S.D. Pohekar). 0360-5442/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.energy.2005.04.012
1216
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
cooking needs. The consumption is increasing due to population growth. This has led to environmental problems associated with deforestation and fuel-wood shortage. Solar cookers and their variations, such as, PSC are low cost options for meeting the cooking energy needs as well as environmental protection. The PSC is an emerging device which has a great potential in India. However, PSC technology will have to compete with prevalent cooking devices in the country. Even though over 5,00,000 SBCs have been sold so far, the large potential of solar cookers is yet untapped [2]. Limited outreach of the technology is attributed to technical limitations, costs, awareness, several behavioral issues, distribution and service network, etc. The Ministry of Non-conventional Energy Sources (MNES), Government of India has devised innovative schemes for promoting the use of PSC. Success of such schemes will be determined by future acceptance of these technologies by the users. Utility assessment is likely to give feedback to the planners and policy makers for formulating strategies for further commercialization. In this paper, the utility of PSC in India with respect to eight prevalent cooking devices has been computed by knowing users’ preferences and expert opinion on thirty different criteria and additive MAUT framework.
2. Prevailing evaluation practices The history of solar cookers goes back a long time. The first reference to solar cooking was that of Nicholas-de-Saussure (1740–1799). There are 50 major designs of solar cookers and more than 100 variations, of which many are already patented [3]. The evaluation of solar cookers has been attempted by various researchers since 1960. In 1961, Volunteers for Technical Assistance (VITA) attempted a detailed review of solar cookers considering such criteria as cooking performance, durability, cost, weight, portability, ease of operation, ease of manufacture and adoptability to local skills and materials [4]. Technical evaluation of various solar cooker designs is attempted by various authors [5,6]. Some of the studies attempted to assess the marketability of SBCs. Amongst these, the review by Gujarat Energy Development Agency (GEDA) [7] is important. Bowman and Blatt [8] attempted the assessment of various solar cookers considering various techno-economic-psychological parameters on 100 point scale. Sharan and Naik [9] attempted to discover the socio-psychological factors determining the acceptance of SBCs in India. The PSC has been introduced in Indian climatic conditions in 1999. The technical evaluation conducted at Technical Backup Units (TBUs) for solar thermal devices for MNES [10] have reported their overall performance to be satisfactory. The present evaluation attempts to quantify the comprehensive utility of PSC vis-a`-vis other cooking devices. To this effect, various criteria and devices are identified. The prevailing cooking devices are assessed with these criteria using the additive MAUT framework.
3. Selection of criteria Cooking energy substitution by renewables is a burning issue and is discussed widely in the literature. Acceptance studies on solar cookers in various parts of India, diagnostic case studies for solar cookers, conjoint analysis for solar cookers and experiences of energy substitution in other developing economies
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1217
are studied to understand end users feedback and simultaneously identify the criteria for the present evaluation. Table 1 presents a summary of 30 criteria and their significance in evaluation of a cooking device in general and PSC in particular discussed in the above studies. Major aspects of assessment are classified as technical, economic, social/environmental, behavioral, and commercial. Fig. 1 presents twenty one criteria of qualitative and nine of quantitative nature. Fuel consumption, cooking time, durability, size/weight/space needs and various costs involved are identified to be quantitative in nature.
4. Selection of devices A large variety of cooking devices are used in India based on prevailing climatic conditions and the socio-economic settings. Traditional low cost devices like chulhas (cook-stoves) are widely used by rural masses which constitute about 68% of the total population. These devices have very low thermal efficiencies (10–15%) and are highly polluting. Also, there are operational difficulties faced by rural women. With the increasing scarcity of fuel-wood, rural women have to walk up to 10 km and spend 3–4 h a day for its collection. Improved chulhas have durability of 2–4 years and require fuel-wood in lesser quantity. However, of 23 Million improved chulhas installed in India, only 6 Million are found to be functional [20]. Biogas plants need higher initial investment and trained manpower for installation. Many biogas plants are non-functional due to non-availability of water throughout the year. The problems identified for limited use of improved chulhas and biogas stoves are operational, social and behavioral leading to non-participation of masses. Kerosene stoves have good thermal efficiency, benefits of simplicity and availability due to good market network and hence are used in semi-urban areas. Due to high costs and weak supply chain in rural India, only 1.3% of rural houses use LPG compared with 27.2% in urban homes [21]. There is pressure to increase the subsidy of kerosene and LPG to further reduce the prices. The subsidies are posing a heavy burden on overall economy of the country. Modern devices such as micro-wave ovens and electric ovens are not affordable by the masses due to high capital costs, high operating costs and intermittent electrical supply. In view of the above problems, solar cooking can be a sustainable energy option for cooking needs. However, in the foreseeable future, PSC technology will have to compete with conventional cooking energy technologies. The PSC considered for evaluation in the present review can deliver 0.6 kW of thermal power with a thermal efficiency of 55–60% and stagnation temperature of 350–400 8C [10]. Thus the cooker is useful for boiling, roasting, frying and baking operations. The cooker is user friendly, convenient and can meet the needs of around 15 people. Community versions of PSC with 7–9 m2 of reflector area and an average efficiency of 25–30% are also installed in India.
5. Multi-criteria evaluation Since majority of the criteria are qualitative in nature, the evaluation on these criteria is conducted through a survey. Since there are 30 criteria involved in the problem, two questionnaires in the form of tabular columns are found to be suitable; one for recording the importance of criteria and the other for recording the performance of identified devices. The questionnaires are tested, validated, modified and opinion is collected from thirty experts. Since there are a wide range issues, from technical
1218
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
Table 1 Significant criterion in cooking energy decisions Criteria no and name
Desired value
Significance
References
CR 1. Fuel consumption CR 2. Cooking timea CR 3. Durabilitya CR 4. Quality, reliability CR 5. Sophistication level CR 6. Size/Weight/Space needsa CR 7. Ruggedness CR 8. Continuity of use CR 9. Need for tracking CR 10. Nutrition value of food CR 11. Initial costa
Low Low High High High Low
Reduced cost, Reduced fuel collection time Higher efficiency, More convenience, Lower nutrient value Reduced costs, Higher reliability, Lower maintenance Lower maintenance Better control of heat, convenience Increased suitability, lower bulk of the device
[1,11] [12,13] [9,11] [9,12] [14,15] [9,11]
High High Low High
Lower maintenance Reduced need of additional system, More convenience Only for solar cookers, Efficiency, Reduced cooking time Healthy food
[9,11] [11,17] [11,16] [13]
Low
[11,13]
CR 12. Fuel cost per montha
Low
CR 13. Maintenance costa
Low
CR 14. Available subsidya CR 15. Rate of interest on loan if anya CR 16. Pollution hazards CR 17. Human drudgery
Low Low
Easy commercialization, Increased Reliability, Convenience Easy commercialization, Increased Reliability, Convenience Easy commercialization, Increased Reliability, Convenience Reduced quality, Easy commercialization Easy commercialization
[1,17,19] [1,19]
High High High High High High High Low
Increased social utility Convenience, Increased productive time for earning for family, education etc Increased social utility Increased commercialization, Higher motivation to use Increased commercialization Increased suitability Versatile cooking, Convenience, Higher motivation to use Versatile cooking, Convenience, Higher motivation to use Versatile cooking, Convenience, Higher motivation to use Convenience, Reduced Costs, Increased Reliability
High
Commercialization, Convenience, Higher motivation to use
[15,19]
High
Increased reliability, Commercialization
[15,18]
High High Low
Better commercialization Better commercialization Convenience, Higher motivation to use
[18] [18] [15,16]
a
CR 18. Overall safety CR 19. Aesthetics CR 20. Motivation to buy CR 21. Taste of food CR 22. Cleanliness of utensils CR 23. Ease of operation CR 24. Type of dishes cooked CR 25. Need for additional cooking system CR 26. Improvement in models CR 27. Spares and after sales service CR 28. Distribution network CR 29. Market research CR 30. Need for user training a
Low Low
[11,13] [11,13] [11,13] [15,18]
[1,17,19] [9,18] [18] [11,14] [11,14] [11,14] [11,14] [14,15]
Quantitative criteria.
improvements to need for user training, experts are identified from various groups. The judgments of energy educators as awareness creators (10), policy makers (10), and researchers as technology up graders (10) are collected though postal, e-mail and personal interactions. The preferences of actual users of the identified devices (30) also are considered in the study. Assessment of scaling constants is
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1219
Fig. 1. Classification of criteria.
determined by collecting the judgments of thirty experts on a 10 point qualitative linear scale (10 representing very high importance and 2 representing very low importance). Weighted average scores are computed for importance of criteria by using a custom built interface developed in a Visual Basic environment. The scaling constants indicate weighted average importance of criteria and are calculated as follows. Pj Scaling$constant Z
i
Wj Nj N
(1)
1220
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
Where Wj indicates the importance of jth criteria (on assigned scale) by the numbers of respondents Nj for that criteria and N are total numbers of responses. The judgments of experts and users are found to be consistent with each other and are indicated in last column of Table 2. The present analyses use the aggregate opinion of all the respondents. The evaluation matrix (Table 2) also is formulated by collecting opinions from experts, users and the published literature. Qualitative criteria judgments are aggregate based on opinion of all the respondents. The values for quantitative criteria are directly taken from published literature [1,10, 21–23]. The values of 2, 4, 6, 8 and 10 are allotted to indicate values of performance of a device as very low, low, moderate, high and very high, respectively. For example, need of user training is very
Table 2 Evaluation matrix for all criteria Device no criteria
Chulha (1)
Improved chulha (2)
Kerosene stove (3)
Biogas stove (4)
LPG stove (5)
Micro wave oven (6)
Electric oven (7)
Solar box cooker (8)
Parabolic solar cooker (9)
Scaling constant
CR 1.a CR 2.a CR 3.a CR 4. CR 5. CR 6.a CR 7. CR 8. CR 9. CR 10. CR 11.a CR 12.a CR 13.a CR 14.a CR 15.a CR 16. CR 17. CR 18. CR 19. CR 20. CR 21. CR 22. CR 23. CR 24. CR 25. CR 26. CR 27. CR 28. CR 29. CR 30.
2 60 1 2 2 2 2 10 0 6 10 20 0 0 0 10 10 2 2 2 6 2 10 10 2 2 10 10 2 2
1 60 4 2 2 1 2 10 0 6 50 10 0 50 0 10 10 2 2 2 6 2 10 10 2 2 10 6 2 2
0.5 30 15 8 6 2 6 10 0 6 200 100 50 0 0 8 6 4 4 6 2 2 10 10 4 6 10 6 8 4
0 15 5 4 6 50 6 4 0 6 5000 0 200 2000 0 2 6 10 4 6 6 8 4 10 4 4 4 6 2 8
0.25 15 20 10 10 10 10 10 0 6 4000 250 50 0 13 4 2 6 10 10 6 10 8 10 6 10 10 2 10 4
2 5 5 10 10 5 8 10 0 2 8000 200 200 0 13 10 2 2 10 8 2 10 4 2 8 10 6 6 10 10
2 30 5 6 10 3 8 10 0 2 5000 400 200 0 13 10 2 2 8 4 10 10 4 2 10 6 4 2 4 6
0 180 10 6 4 5 8 4 4 10 2000 0 50 500 3 0 2 10 8 4 10 10 6 2 10 6 4 2 4 6
0 20 20 6 4 15 8 2 10 8 7000 0 20 2000 3 0 2 8 8 4 10 10 4 4 10 6 2 2 4 8
0.0355 0.0384 0.0397 0.0376 0.0326 0.0332 0.0310 0.0335 0.0326 0.0371 0.0371 0.0293 0.0274 0.0252 0.0230 0.0217 0.0296 0.0433 0.0348 0.0314 0.0381 0.0368 0.0387 0.0332 0.0306 0.0342 0.0352 0.0355 0.0315 0.0323
a
Quantitative criteria.
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1221
low for chulha and hence assigned 10 points, whereas it is very high in case of microwave ovens which is assigned 2 points. Since biogas stove, SBC and PSC are dependent on renewable energy, fuel consumption and fuel costs are taken as zero. On the other hand, need for tracking is assigned zero value in the evaluation matrix for all other devices except SBC and PSC, as the energy input for cooking with these devices is independent of the Sun’s position. Many of the cooking devices require lower capital investments and are not given any subsidy. The relevant score assigned is zero for these devices. The quantitative scores in evaluation matrix indicate quantity of fuel consumed per family per day in kilograms, cooking time in minutes, durability in years and various costs in Rupees. Size/weight /space needs are indicated by the overall bulk in kilograms whereas rate of interest on loan is indicated in percent (Table 3). Table 3 Normalized evaluation matrix Device no criteria
Chulha (1)
Improved chulha (2)
Kerosene stove (3)
Biogas stove (4)
LPG stove (5)
Micro wave oven (6)
Electric oven (7)
Solar box cooker (8)
Parabolic solar cooker (9)
CR 1. CR 2. CR 3. CR 4. CR 5. CR 6. CR 7. CR 8. CR 9. CR 10. CR 11. CR 12. CR 13. CR 14. CR 15. CR 16. CR 17. CR 18. CR 19. CR 20. CR 21. CR 22. CR 23. CR 24. CR 25. CR 26. CR 27. CR 28. CR 29. CR 30.
0.0000 0.6857 0.0000 0.0000 0.0000 0.9796 0.0000 1.0000 1.0000 0.5000 1.0000 0.9500 1.0000 1.0000 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.5000 0.0000 1.0000 1.0000 1.0000 0.0000 1.0000 1.0000 0.0000 1.0000
0.5000 0.6857 0.1579 0.0000 0.0000 1.0000 0.0000 1.0000 1.0000 0.5000 0.9950 0.9750 1.0000 0.9800 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.5000 0.0000 1.0000 1.0000 1.0000 0.0000 1.0000 0.5000 0.0000 1.0000
0.7500 0.8571 0.7368 0.7500 0.5000 0.9796 0.5000 1.0000 1.0000 0.5000 0.9762 0.7500 0.7500 1.0000 1.0000 0.2000 0.5000 0.2500 0.2500 0.5000 0.0000 0.0000 1.0000 1.0000 0.7500 0.5000 1.0000 0.5000 0.7500 0.7500
1.0000 0.9429 0.2105 0.2500 0.5000 0.0000 0.5000 0.2500 1.0000 0.5000 0.3755 1.0000 0.0000 1.0000 0.7692 0.8000 0.5000 1.0000 0.2500 0.5000 0.5000 0.7500 0.0000 1.0000 0.7500 0.2500 0.2500 0.5000 0.0000 0.2500
0.8500 0.9429 1.0000 1.0000 1.0000 0.8163 1.0000 1.0000 1.0000 0.5000 0.5006 0.3750 0.7500 0.0000 0.0000 0.6000 1.0000 0.5000 1.0000 1.0000 0.5000 1.0000 0.7500 1.0000 0.5000 1.0000 1.0000 0.0000 1.0000 0.7500
0.0000 1.0000 0.2105 1.0000 1.0000 0.9184 0.7500 1.0000 1.0000 0.0000 0.0000 0.5000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 1.0000 0.7500 0.0000 1.0000 0.0000 0.0000 0.2500 1.0000 0.5000 0.5000 1.0000 0.0000
0.0000 0.8571 0.2105 0.5000 1.0000 0.9592 0.7500 1.0000 1.0000 0.0000 0.3755 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 0.7500 0.2500 1.0000 1.0000 0.0000 0.0000 0.0000 0.5000 0.2500 0.0000 0.2500 0.5000
1.0000 0.0000 0.4737 0.5000 0.2500 0.9184 0.7500 0.2500 0.6000 1.0000 0.7509 1.0000 0.7500 0.7500 0.7692 1.0000 1.0000 1.0000 0.7500 0.2500 1.0000 1.0000 0.3300 0.0000 0.0000 0.5000 0.2500 0.0000 0.2500 0.5000
1.0000 0.9143 1.0000 0.5000 0.2500 0.7143 0.7500 0.0000 0.0000 0.7500 0.1252 1.0000 0.9000 1.0000 0.7692 1.0000 1.0000 0.7500 0.7500 0.2500 1.0000 1.0000 0.0000 0.2500 0.0000 0.5000 0.0000 0.0000 0.2500 0.7500
1222
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
6. The additive MAUT model There are various methods to assess the preferences, such as Analytic Hierarchy Process (AHP), Simple Multi-Attribute Rating Technique (SMART), etc. While SMART is a simple method for assessing preferences, it has limitations in handling inconsistent units. The AHP model involves higher complexity with 30 criteria and consistency checks at each level. The advantage of the additive MAUT is its simplicity. It combines the main advantages of simple scoring techniques and optimization models. Utility is a measure of desirability or satisfaction and provides a uniform scale to compare and/or combine tangible and intangible criteria. The MAUT theory is developed to help decision makers assign utility values to devices in terms of single attribute utility functions and combine individual evaluations to obtain overall utility values. It considers the decision maker’s preferences in the form of a utility function which is defined over a set of attributes. However, it is observed that MAUT is not very extensively used in energy planning [24]. This may be due to requirements of interactive decision environment required for formulating utility functions and complexity of computing scaling constants using the MAUT algorithm. Selecting portfolios for solar energy projects [25] and energy policy making [26] are the applications identified in the literature. The utility value can be determined by the following three steps: (i) Determination of single attribute utility functions. The values of utilities vary between zero and one and reflect the level of importance in the achievement of that attribute. (ii) Verification of preferential and utility independence conditions. (iii) Derivations of the multi attribute utility function to determine intermediate utilities. All decisions involve choosing one or a few, from several, alternatives. Each alternative is assessed for desirability on a number of scored criteria. Utility function connects the criteria scores with desirability. The utility functions are of two important forms, viz. additive and multiplicative. The most common formulation of a multi-criteria utility function is the additive model. Ui O Wj ; uij ;
for all i;
(2)
where Ui is the overall utility value of device i uij is the utility value of the jth criterion for the ith device Uij equals u (Xi), for 1RiRn and iRjRm Xi equals (xij) for 1RiRn and 1RjRm Xi designates a specific value of xij n is total number of criteria, m is total number of devices Wj is the scaling constant of jth criteria
7. Results and discussions An analysis of scaling constants indicate that utility of the selected cooking energy devices is governed by technical (0.352) followed by behavioral (0.246), commercial (0.167) and economic criteria (0.141). The sum of all scaling constants is unity. It also is observed that social/environmental
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1223
criteria are assigned lower importance (0.095), overall safety (0.0433), durability (0.0397) and ease of operation (0.0387) also received low importance ranking by the respondents. Pollution hazard (0.0217) is ranked the least important. An analysis of the evaluation matrix reveals the strengths of PSC as fuel consumption, durability, nutritional value of food, fuel cost, available subsidy, pollution hazards, human drudgery, taste of food, cleanliness of utensils etc. The SBCs are also indicated to have similar strengths, except cooking time. The LPG stove has higher values for all the criterion except initial, fuel cost, rate of interest on loan, distribution network. Values for chulha and improved chulha indicate strengths in terms of continuity of use, ease of operation, type of dishes cooked, spares and after sales service. Kerosene stoves have moderate values for most of the criteria. Micro-wave and electric ovens have strengths for many criteria except costs and need for user training. Following the formulation of the evaluation matrix, the next step is identification of the best and worst outcomes (criteria scores) for each of the criteria. The best outcome for a particular criterion is assigned utility value of unity and the worst as zero. To assign intermediate values, utility functions are developed by the technique known as standard gambling [27]. The decision maker is offered certain outcome with probability pZ1. Risk option with a probabilistic outcome in the form of gamble with best outcome considering probability p or worst outcome with probability of 1-p is also offered to the experts. Fig. 2 shows the pair of lotteries for criterion ‘motivation to buy (CR 20)’. Lottery 1 and Lottery 2 shows best consequences of the same criterion with certain probability of p and worst probability of 1-p respectively. The decision maker in the present case selected the indifference probability of pZ0.5 after mental trial and error. This indifference allows assessing the utility values from the principle of expected utility from the probability theory. For all the criteria, linear utility functions are considered to assess the utilities. There are two kinds of criteria in the present problem viz; the maximization criteria (the maximum value is desirable e g. durability) and the minimization criteria (the minimization value is desirable, e g. initial cost). Since the present problem uses probability of 0.5 and linear utility functions, the intermediate utility values are thus obtained by normalizing the evaluation matrix as follows.
uðjÞ Z
Aj K Amin Amax K Amin
For maximization criteria
Fig. 2. Pair of Lotteries for motivation to buy (CR 20).
(3)
1224
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
uðjÞ Z
Amax K Aj Amax K Amin
For minimization criteria
(4)
Where Aj represents the score assigned to cooking device in evaluation matrix. Amax and Amin are the maximum and minimum scores assigned for the selected criteria for identified devices. Following the above normalization of evaluation matrix is computed. The final utilities are computed by multiplying the normalized utility values (on the scale of zero to one) by respective scaling constants. The final utility Table 4 Utility values for identified devices Device no criteria
Chulha (1)
Improved chulha (2)
Kerosene stove (3)
Biogas stove (4)
LPG stove (5)
Micro wave oven (6)
Electric oven (7)
Solar box cooker (8)
Parabolic solar cooker (9)
CR 1. CR 2. CR 3. CR 4. CR 5. CR 6. CR 7. CR 8. CR 9. CR 10. CR 11. CR 12. CR 13. CR 14. CR 15. CR 16. CR 17. CR 18. CR 19. CR 20. CR 21. CR 22. CR 23. CR 24. CR 25. CR 26. CR 27. CR 28. CR 29. CR 30. Additive utility Ranking
0.0000 0.0263 0.0000 0.0000 0.0000 0.0325 0.0000 0.0335 0.0326 0.0186 0.0371 0.0278 0.0274 0.0252 0.0230 0.0000 0.0000 0.0000 0.0000 0.0000 0.0191 0.0000 0.0387 0.0332 0.0306 0.0000 0.0352 0.0355 0.0000 0.0323 0.5086
0.0178 0.0263 0.0063 0.0000 0.0000 0.0332 0.0000 0.0335 0.0326 0.0186 0.0369 0.0286 0.0274 0.0247 0.0230 0.0000 0.0000 0.0000 0.0000 0.0000 0.0191 0.0000 0.0387 0.0332 0.0306 0.0000 0.0352 0.0178 0.0000 0.0323 0.5156
0.0266 0.0329 0.0293 0.0282 0.0163 0.0325 0.0155 0.0335 0.0326 0.0186 0.0362 0.0220 0.0206 0.0252 0.0230 0.0043 0.0148 0.0108 0.0087 0.0157 0.0000 0.0000 0.0387 0.0332 0.0230 0.0171 0.0352 0.0178 0.0236 0.0242 0.6600
0.0355 0.0362 0.0084 0.0094 0.0163 0.0000 0.0155 0.0084 0.0326 0.0186 0.0139 0.0293 0.0000 0.0252 0.0177 0.0174 0.0148 0.0433 0.0087 0.0157 0.0191 0.0276 0.0000 0.0332 0.0230 0.0086 0.0088 0.0178 0.0000 0.0081 0.5127
0.0302 0.0362 0.0397 0.0376 0.0326 0.0271 0.0310 0.0335 0.0326 0.0186 0.0186 0.0110 0.0206 0.0000 0.0000 0.0130 0.0296 0.0217 0.0348 0.0314 0.0191 0.0368 0.0290 0.0332 0.0153 0.0342 0.0352 0.0000 0.0315 0.0242 0.7581
0.0000 0.0384 0.0084 0.0376 0.0326 0.0305 0.0233 0.0335 0.0326 0.0000 0.0000 0.0147 0.0000 0.0000 0.0000 0.0000 0.0296 0.0000 0.0348 0.0236 0.0000 0.0368 0.0000 0.0000 0.0077 0.0342 0.0176 0.0178 0.0315 0.0000 0.4849
0.0000 0.0329 0.0084 0.0188 0.0326 0.0318 0.0233 0.0335 0.0326 0.0000 0.0139 0.0000 0.0000 0.0000 0.0000 0.0000 0.0296 0.0000 0.0261 0.0079 0.0381 0.0368 0.0000 0.0000 0.0000 0.0171 0.0088 0.0000 0.0079 0.0162 0.4162
0.0355 0.0000 0.0188 0.0188 0.0082 0.0305 0.0233 0.0084 0.0196 0.0371 0.0279 0.0293 0.0206 0.0189 0.0177 0.0217 0.0296 0.0433 0.0261 0.0079 0.0381 0.0368 0.0128 0.0000 0.0000 0.0171 0.0088 0.0000 0.0079 0.0162 0.5805
0.0355 0.0351 0.0397 0.0188 0.0082 0.0237 0.0233 0.0000 0.0000 0.0278 0.0046 0.0293 0.0247 0.0252 0.0177 0.0217 0.0296 0.0325 0.0261 0.0079 0.0381 0.0368 0.0000 0.0083 0.0000 0.0171 0.0000 0.0000 0.0079 0.0242 0.5637
7
5
2
6
1
8
9
3
4
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1225
value is found by adding the utilities of devices on different criteria as follows. Utility Z
30 X
Wj uij
(5)
1
Table 4 shows utility values and ranking for identified devices following this methodology. The utility of a LPG stove is found to be the highest at 0.7581 followed by kerosene stove (0.6600), SBC (0.5805) and PSC (0.5637) on a scale between 0 and 1. An electric stove has the least utility of 0.4162. The LPG stove is a preferred device because of its strengths on most parameters. The strengths of PSC and SBC are on environmental and social issues for which the scaling constants are the smallest. Sensitivity analyses are conducted to formulate strategies for increasing the utility of PSC (Table 5). The utility is sensitive to technical, as well as behavioral issues. Alteration of economic criteria does not have much effect on the utility. The problem of dissemination appears to be sensitive to behavioral issues such as type of dishes cooked and convenience. Certain commercial parameters such as availability of spares and after sales service and improvement in existing models are also found to be key factors in increasing the overall utility of PSC in the Indian domestic sector. The potential strategies for better dissemination may be similar for PSC and SBC. Reduction in size/weight/space improves the utility value. Making PSC a trendy and portable device can be achieved by using aluminium, fiber composites to mount reflectors and caster wheels for easy movement. These measures however do not make any significant change in ranking of PSC. Improvements in quality/reliability of the product can be a good alternative strategy. Better quality systems however warrant large volumes of production to be cost effective. A competing device using that strategy is the kerosene stove. Attempts have been made to check the sensitivity to a few behavioral criteria, viz. motivating potential buyers, maximizing ease of operation and type of dishes cooked. The studies indicate changes in the ranking pattern. This can be achieved by creating awareness amongst the potential users, and arranging cooking demonstrations etc. Addressing all behavioral issues simultaneously may make substantial change in the ranking pattern. Table 5 Rankings of PSC for alternative strategies No
Alternative strategy
Criteria no
Utility
Rank
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
Reducing the bulk weight, size etc Improving aesthetics Improvement in existing models Improving quality and reliability Extensive market research Reducing cost to Rs.2000 Motivating potential buyers Cooking demos for type of dishes cooked Improving after sales service Improvements in distribution network Maximizing ease of operation Improving on all the technical criteria Improving on all the commercial criteria Improving on all the behavioral criteria Improvements on all criteria
6 19 26 4 29 11 20 24 27 28 23 1–10 26–30 19–25 1–30
0.5427 0.5724 0.5808 0.5825 0.5837 0.5869 0.5872 0.5886 0.5989 0.5992 0.6024 0.6642 0.6832 0.6901 1.0000
4 4 3 3 3 3 3 3 3 3 3 2 2 2 1
1226
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
Commercial criteria like improvements in aesthetics, improvements in existing models, extensive market research and good after sales service can improve the ranking pattern.
8. Conclusions There is a need to assess the utility of cooking devices by considering multiple criteria in view of prevailing techno-economic and socio-commercial scenarios. Additive MAUT provides one such framework for ranking cooking devices by considering quantitative and qualitative criteria. The utility value can be used as a yardstick for knowing perceptions of different groups involved in the decision process. The utility values obtained reveal that LPG has the highest and electric stove has the lowest utility. The SBCs occupy third position in the ladder followed by PSC. The sensitivity analyses provide better insight into the problem of lower utility of PSC in the present Indian context. The kerosene stove may emerge as a competitive option to LPG and PSC due to lower fuel needs, better cooking time, a good market and a service network supported by economics. Sensitivity analysis indicates that the problem of better dissemination of PSC is merely not an economic issue. The utility of PSC on technical, behavioral and commercial aspects should be improved. Creating awareness among the masses, building confidence based on behavioral issues, together with technical improvements in the product can be taken into account in formulating the strategies for better dissemination.
References [1] Reddy BS. Overcoming the energy efficiency gap in India’s household sector. Energy Policy 2003;31:1117–27. [2] Ministry of Non-conventional Energy Sources. Annual Report, New Delhi: Government of India. CGO Complex Lodi Road; 2003. [3] Kundapur A. Reviews of solar cooker designs. TERI Inf Digest Energy 1998;8(1):1–37. [4] Volunteers in Technical Assistance. Evaluation of Solar Cookers. VITA report. No 10. 1600 Wilson Boulevard, Suite 1030, Arlington, Virginia; 1961. [5] Das TCT, Karmarkar S, Rao DP. Solar box cooker. 1. Modeling. Solar Energy 1994;52(3):265–72. [6] Funk PA, Larson DL. Parametric model of solar cooker performance. Solar Energy 1998;62(1):63–8. [7] Gujarat Energy Development Agency. Proposal for Popularizing Solar Cookers. Vadodara: II, Floor Suraj Plaza II Sayaji Ganj; 1979. [8] Bowman TE, Blatt JH. Solar cookers, history, design, fabrication, testing and evaluation. Florida: Florida Institute of Technology; 1978. [9] Sharan Girija, Naik G. Assessing consumer preference for product features selection: solar cookers. Vikalpa 1997;22(4): 49–54. [10] Sawhney RL, Buddhi D, Gautam RP editors. Renewable energies and energy efficiency for sustainable development. Proceedings of 23rd National Renewable Energy Convention. Indore, India. Solar Energy Society of India, New Delhi; 1999. [11] Gore AP, Paranjape SA, Pandit SJ, Prayag V. Why solar cookers do not sell. Changing Villages 1990;9(4):219–25. [12] Ahmad B. Users and disusers of box solar cookers in urban India-implications for solar cooking projects. Solar Energy 2000;69(1–6 Suppl):209–15. [13] Kumar S, Kandpal TC, Mullik SC. Solar cooker use in Delhi: Pilot survey of urban households. Int J Ambient Energy 1997;18(2):77–82.
S.D. Pohekar, M. Ramachandran / Energy 31 (2006) 1215–1227
1227
[14] Avinashilingam Institute for Home Science and Higher Education for Women. Proceedings of 3rd International Conference on Solar Cookers Use and Technology, Coimbatore, India; 1997. [15] Moulik R. Socio-psychological and economic factors affecting acceptability of solar cookers in Gujarat. Unpublished Thesis. Ahmedabad: Indian Institute of Management; 1985. [16] Sama Shrestha. Teaching and training communities of Kathmandu valley in the use of solar parabolic cookers. In: Misra Anil, editor. Renewable energy education current practices and future projections. New Delhi: Tata Energy Research Institute; 1998. p. 133–43. [17] Philip SK, Makwana HM, Singhal AK. Monitoring of subsidized solar cookers in Gujarat: a case study. SESI J 1987;1: 37–43. [18] Mohanty RC. Gujarat energy development agency: the case of solar cookers. Vikalpa 1991;16(2):65–79. [19] Biermann E, Grupp M, Palmer R. Solar cooker acceptance in South Africa: results of comparative field tests. Solar Energy 1999;66(6):401–7. [20] Neudoerfler RC, Malhotra Preeti, Ramana PV. Participatory rural energy planning in India—a policy context. Energy Policy 2001;29:371–81. [21] Tata Energy Data and Directory Yearbook (TEDDY) 2001/02. New Delhi: Tata Energy Research Institute; 2002. [22] Gupta S, Ravindranath NH. Financial analysis of cooking energy options for India. Energy Convers Manage 1997;38(10): 1869–77. [23] Ramanathan R, Ganesh LS. Multi-objective analysis of cooking-energy alternatives. Energy 1994;19(4):469–78. [24] Huang JP, Poh KL, Ang BW. Decision analysis in energy and environmental modeling. Energy 1995;20(9):843–55. [25] Golabi K, Kirkwood CW, Sicherman A. Selecting a portfolio of solar energy projects using multiattribute preference theory. Manage Sci 1981;22(2):174–89. [26] Jones M, Hope C, Hughes R. A multi-attribute value model for the study of UK energy policy. J Operations Res Soc 1990; 41(10):919–29. [27] Keeney RL, Raiffa H. Decisions with multiple objectives: preferences and value tradeoffs. Cambridge: Cambridge University Press; 1993.