Presentation is loading. Please wait.

Presentation is loading. Please wait.

IME634: Management Decision Analysis

Similar presentations


Presentation on theme: "IME634: Management Decision Analysis"โ€” Presentation transcript:

1 IME634: Management Decision Analysis
Raghu Nandan Sengupta Industrial & Management Department Indian Institute of Technology Kanpur TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

2 RNSengupta,IME Dept.,IIT Kanpur,INDIA
Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method was developed for the INTEGRated Human Exploration Mission Simulation FaciliTY (INTEGRITY) project in the Johnson Space Centre to assess the priority of a set of human spaceflight mission simulators TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

3 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS (contdโ€ฆ) One assumes utility function is monotonic, in the sense the more/less you get more/less you want The basic premise based on which TOPSIS works is the fact that selected alternatives should have the shortest distance from the positive-ideal solution, and the farthest distance from the negative-ideal solution TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

4 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS (contdโ€ฆ) Choose positive ideal solution (PIS) of the original ranking problem Choose negative ideal solution (NIS) of the original ranking problem Find distances from each decisions/alternatives, ๐ด ๐‘– , ๐‘–= 1,โ‹ฏ,๐‘š to PIS, which is given as ๐‘‘ ๐ด ๐‘– ,๐‘ƒ๐ผ๐‘† Find distances from each decisions/alternatives, ๐ด ๐‘– , ๐‘–= 1,โ‹ฏ,๐‘š to NIS which is given as ๐‘‘ ๐ด ๐‘– ,๐‘๐ผ๐‘† TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

5 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS (contdโ€ฆ) Euclidean distance measure is used and we ensure our main motivation is to minimize the dispersion Calculate ๐‘Ÿ ๐‘– = ๐‘‘ ๐ด ๐‘– ,๐‘๐ผ๐‘† ๐‘‘ ๐ด ๐‘– ,๐‘๐ผ๐‘† +๐‘‘ ๐ด ๐‘– ,๐‘ƒ๐ผ๐‘† = ๐ด ๐‘– โˆ’๐‘๐ผ๐‘† ๐ด ๐‘– โˆ’๐‘๐ผ๐‘† ๐ด ๐‘– โˆ’๐‘ƒ๐ผ๐‘† 2 The basic premise being Euclidean distance portrays the concept of utility function, ๐‘ˆ ๐‘Š which is quadratic TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

6 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS (contdโ€ฆ) Minimizing ๐‘ˆ ๐‘Š ensures minimizing dispersion, i.e., minimization of ๐‘‰๐‘Ž๐‘Ÿ ๐‘‹ One ranks the ratios, ๐‘Ÿ ๐‘– , to get the best alternative TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

7 RNSengupta,IME Dept.,IIT Kanpur,INDIA
Algorithm for TOPSIS Assume decisions/alternatives as ๐ด ๐‘– , ๐‘–= 1,โ‹ฏ,๐‘š Assume attributes/decision criteria/goals are ๐ถ ๐‘— , ๐‘—=1,โ‹ฏ,๐‘› We state the pseudo-codes for the working principle of TOPSIS TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

8 Algorithm for TOPSIS (contdโ€ฆ)
1: DEFINE: ๐‘ฟ ๐‘šร—๐‘› (matrix consisting of priority scores assigned to decisions/alternatives), ๐ด ๐‘– ,based on attributes/decision criteria/goals, ๐ถ ๐‘— ; ๐‘ค ๐‘— (weight for the attributes/decision criteria/goals) such that ๐‘—=1 ๐‘› ๐‘ค ๐‘— =1;๐‘ฉ (benefit matrix);๐‘ช (cost matrix); ๐‘Ÿ ๐‘–,๐‘— = ๐‘ฅ ๐‘–,๐‘— ๐‘–=1 ๐‘š ๐‘ฅ ๐‘–,๐‘— 2 ; ๐ด๐ผ๐‘†= ๐‘ฃ 1 + ,โ‹ฏ, ๐‘ฃ ๐‘š + = max โˆ€๐‘– ๐‘ฃ ๐‘–,๐‘— ๐‘—โˆˆ๐‘ฉ , min โˆ€๐‘– ๐‘ฃ ๐‘–,๐‘— ๐‘—โˆˆ๐‘ช (negative ideal solution);๐‘ƒ๐ผ๐‘†= ๐‘ฃ 1 โˆ’ ,โ‹ฏ, ๐‘ฃ ๐‘š โˆ’ = min โˆ€๐‘– ๐‘ฃ ๐‘–,๐‘— ๐‘—โˆˆ๐‘ฉ , max โˆ€๐‘– ๐‘ฃ ๐‘–,๐‘— ๐‘—โˆˆ๐‘ช (positive ideal solution); ๐‘† ๐‘– + = ๐‘—=1 ๐‘› ๐‘ฃ ๐‘–,๐‘— โˆ’ ๐‘ฃ ๐‘— + 2 ; ๐‘† ๐‘– โˆ’ = ๐‘—=1 ๐‘› ๐‘ฃ ๐‘–,๐‘— โˆ’ ๐‘ฃ ๐‘— โˆ’ 2 ; ๐‘‡ ๐‘– = ๐‘† ๐‘– โˆ’ ๐‘† ๐‘– + + ๐‘† ๐‘– โˆ’ (relative closeness);๐‘€= ๐‘† ๐‘– + , ๐‘† ๐‘– โˆ’ (separation measure). Here ๐‘–=1,โ‹ฏ,๐‘š and ๐‘—=1,โ‹ฏ,๐‘› 2: INPUT: ๐‘ฟ ๐‘šร—๐‘› (matrix consisting of priority scores assigned to decisions/alternatives), ๐ด ๐‘– ,based on attributes/decision criteria/goals, ๐ถ ๐‘— ; ๐‘ค ๐‘— (weight for the attributes/decision criteria/goals) such that ๐‘—=1 ๐‘› ๐‘ค ๐‘— =1;๐‘ฉ (benefit matrix);๐‘ช (cost matrix). Here ๐‘–=1,โ‹ฏ,๐‘š and ๐‘—=1,โ‹ฏ,๐‘› 3: START if: ๐‘–=1:๐‘š 4: START if: ๐‘—=1:๐‘› 5: CALCULATE: ๐‘Ÿ ๐‘–,๐‘— = ๐‘ฅ ๐‘–,๐‘— ๐‘–=1 ๐‘š ๐‘ฅ ๐‘–,๐‘— 2 ; ๐‘ฃ ๐‘–,๐‘— = ๐‘ค ๐‘— ๐‘Ÿ ๐‘–,๐‘— where ๐‘–=1,โ‹ฏ,๐‘š and ๐‘—=1,โ‹ฏ,๐‘› 6: END if 7: END if 8: CALCULATE: ๐‘ฃ ๐‘– + ; ๐‘ฃ ๐‘– โˆ’ ;๐ด๐ผ๐‘†;๐‘ƒ๐ผ๐‘†; ๐‘† ๐‘– + ; ๐‘† ๐‘– โˆ’ ; ๐‘€; ๐‘‡ ๐‘– 9: REPORT: ๐ด๐ผ๐‘†;๐‘ƒ๐ผ๐‘†;๐‘€; ๐‘‡ ๐‘– 10: END TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

9 TOPSIS (contdโ€ฆ): Distance measure
The Euclidean distance between vector/points ๐’™= ๐‘ฅ 1 ,โ‹ฏ, ๐‘ฅ ๐‘› and ๐’š= ๐‘ฆ 1 ,โ‹ฏ, ๐‘ฆ ๐‘› is ๐‘–=1 ๐‘› ๐‘ฅ ๐‘– โˆ’ ๐‘ฆ ๐‘– The ๐ฟ 1 norm or Manhattan distance between vector/points ๐’™= ๐‘ฅ 1 ,โ‹ฏ, ๐‘ฅ ๐‘› and ๐’š= ๐‘ฆ 1 ,โ‹ฏ, ๐‘ฆ ๐‘› is ๐‘–=1 ๐‘› ๐‘ฅ ๐‘– โˆ’ ๐‘ฆ ๐‘– . The name relates to the distance a taxi has to drive in a rectangular street grid Mahalanobis distance between random vector/points ๐’™= ๐‘ฅ 1 ,โ‹ฏ, ๐‘ฅ ๐‘› and ๐’š= ๐‘ฆ 1 ,โ‹ฏ, ๐‘ฆ ๐‘› is ๐’™โˆ’๐’š ๐‘‡ ๐‘บ โˆ’1 ๐’™โˆ’๐’š , where ๐‘บ is the covariance matrix TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

10 TOPSIS (contdโ€ฆ): Distance measure
The Hamming distance between vector/points ๐’™= ๐‘ฅ 1 ,โ‹ฏ, ๐‘ฅ ๐‘› and ๐’š= ๐‘ฆ 1 ,โ‹ฏ, ๐‘ฆ ๐‘› is the number of positions at which the corresponding values are different The ๐ฟ โˆž norm between vector/points ๐’™= ๐‘ฅ 1 ,โ‹ฏ, ๐‘ฅ ๐‘› and ๐’š= ๐‘ฆ 1 ,โ‹ฏ,๐‘ฆ is max โˆ€๐‘– ๐‘ฆ ๐‘– โˆ’ ๐‘ฆ ๐‘– The ๐ฟ ๐‘ norm between vector/points ๐’™= ๐‘ฅ 1 ,โ‹ฏ, ๐‘ฅ ๐‘› and ๐’š= ๐‘ฆ 1 ,โ‹ฏ, ๐‘ฆ ๐‘› is ๐‘–=1 ๐‘› ๐‘ฅ ๐‘– โˆ’ ๐‘ฆ ๐‘– ๐‘ ๐‘ TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

11 TOPSIS: Step # 01 (Construct the normalized decision matrix)
Assume the decision matrix, ๐‘ฟ= ๐‘ฅ 11 โ‹ฏ ๐‘ฅ 1๐‘› โ‹ฎ โ‹ฑ โ‹ฎ ๐‘ฅ ๐‘š1 โ‹ฏ ๐‘ฅ ๐‘š๐‘› Convert the entries in X into scaled normalized values, ๐‘Ÿ ๐‘–๐‘— = ๐‘ฅ ๐‘–๐‘— ๐‘˜=1 ๐‘š ๐‘ฅ ๐‘˜๐‘— 2 , which has no dimension Thus we get ๐‘น= ๐‘Ÿ 11 โ‹ฏ ๐‘Ÿ 1๐‘› โ‹ฎ โ‹ฑ โ‹ฎ ๐‘Ÿ ๐‘š1 โ‹ฏ ๐‘Ÿ ๐‘š๐‘› = ๐‘ฅ ๐‘˜=1 ๐‘š ๐‘ฅ ๐‘˜ โ‹ฏ ๐‘ฅ 1๐‘› ๐‘˜=1 ๐‘š ๐‘ฅ ๐‘˜๐‘› โ‹ฎ โ‹ฑ โ‹ฎ ๐‘ฅ ๐‘š1 ๐‘˜=1 ๐‘š ๐‘ฅ ๐‘˜ โ‹ฏ ๐‘ฅ ๐‘š๐‘› ๐‘˜=1 ๐‘š ๐‘ฅ ๐‘˜๐‘› 2 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

12 TOPSIS: Step # 01 (Construct the normalized decision matrix) (contd..)
Assume, ๐‘ฟ= Scale the values using normalization concept, i.e., ๐‘Ÿ ๐‘–๐‘— = ๐‘ฅ ๐‘–๐‘— ๐‘˜=1 ๐‘š ๐‘ฅ ๐‘˜๐‘— 2 (you can use any other concept of utility also) ๐‘น= TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

13 TOPSIS: Step # 01 (Construct the normalized decision matrix) (contd..)
๐‘น= Check each column adds up to 1 as it should be TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

14 TOPSIS: Step # 02 (Construct the weighted normalized decision matrix)
If the decision maker decides on the set of weights, depending on his/her preference, then the weight, ๐‘พ= ๐‘ค 1 โ‹ฏ 0 โ‹ฎ โ‹ฑ โ‹ฎ 0 โ‹ฏ ๐‘ค ๐‘› , such that ๐‘—=1 ๐‘› ๐‘ค = 1 Consider, ๐‘พ= Calculate ๐‘ฝ=๐‘น๐‘พ TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

15 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Step # 02 (Construct the weighted normalized decision matrix) (contd..) Thus ๐‘ฝ=๐‘น๐‘พ= ร— Hence ๐‘ฝ=๐‘น๐‘พ= ๐‘ฃ 11 ๐‘ฃ 12 ๐‘ฃ 13 ๐‘ฃ ๐‘ฃ 21 ๐‘ฃ 22 ๐‘ฃ 23 ๐‘ฃ ๐‘ฃ 31 ๐‘ฃ 32 ๐‘ฃ 33 ๐‘ฃ ๐‘ฃ 41 ๐‘ฃ 42 ๐‘ฃ 43 ๐‘ฃ ๐‘ฃ 51 ๐‘ฃ 52 ๐‘ฃ 53 ๐‘ฃ 54 = TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

16 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Step # 03 (Determine the most positive-ideal, most negative-ideal solutions) Calculate ๐‘ฝ + which is most positive ideal solution Where ๐‘ฝ + = max ๐‘– ๐‘ฃ ๐‘–๐‘— ๐‘—โˆˆ๐ฝ ,๐‘–=1,โ‹ฏ,๐‘š = ๐‘ฃ 1 + , ๐‘ฃ 2 + ,โ‹ฏ, ๐‘ฃ ๐‘› + TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

17 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Step # 03 (Determine the most positive-ideal, most negative-ideal solutions) Calculate ๐‘ฝ โˆ’ which is most negative ideal solution Where ๐‘ฝ โˆ’ = min ๐‘– ๐‘ฃ ๐‘–๐‘— ๐‘—โˆˆ๐ฝ ,๐‘–=1,โ‹ฏ,๐‘š = ๐‘ฃ 1 โˆ’ , ๐‘ฃ 2 โˆ’ ,โ‹ฏ, ๐‘ฃ ๐‘› โˆ’ TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

18 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Step # 04 (Calculate the distance based on most positive-ideal, most negative-ideal solutions) Calculate ๐‘† ๐‘– + based on most positive ideal solution Where ๐‘† ๐‘– + = ๐‘—=1 ๐‘› ๐‘ฃ ๐‘–๐‘— โˆ’ ๐‘ฃ ๐‘— , ๐‘–= 1,โ‹ฏ,๐‘š TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

19 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Step # 04 (Calculate the distance based on most positive-ideal, most negative-ideal solutions) Calculate ๐‘† ๐‘– โˆ’ based on most negative ideal solution Where ๐‘† ๐‘– โˆ’ = ๐‘—=1 ๐‘› ๐‘ฃ ๐‘–๐‘— โˆ’ ๐‘ฃ ๐‘— โˆ’ , ๐‘–= 1,โ‹ฏ,๐‘š TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

20 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Step # 05 (Calculate the relative proximity based on ideal solution) Calculate ๐‘ ๐‘– = ๐‘† ๐‘– โˆ’ ๐‘† ๐‘– โˆ’ + ๐‘† ๐‘– + , ๐‘–= 1,โ‹ฏ,๐‘š TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

21 RNSengupta,IME Dept.,IIT Kanpur,INDIA
TOPSIS: Example Consider the problem related to buying a house/apartment among four (04) choices, where the decision to buy the house/apartment is based on eleven (11) different parameters/criterion which are City Price Loan availability/conditions Location Number of rooms Safety Proximity to markets Proximity to schools Proximity to hospitals Facilities available Resale condition TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

22 TOPSIS: Example (contd..)
๐‘ฟ= Use the normalization formulae as , ๐‘Ÿ ๐‘–๐‘— = ๐‘ฅ ๐‘–๐‘— ๐‘˜=1 ๐‘› ๐‘ฅ ๐‘–๐‘˜ 2 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

23 TOPSIS: Example (contd..)
๐‘น= For example ๐‘Ÿ 3,2 = =0.5964 ๐‘Ÿ 4,8 = =0.6471 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

24 TOPSIS: Example (contd..)
Consider ๐‘พ= (0.119, 0.106, 0.115, 0.090, 0.113, 0.061, 0.064, 0.113, 0.096, 0.065, 0.058), where ๐‘—=1 ๐‘›=11 ๐‘ค ๐‘— = =1 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

25 TOPSIS: Example (contd..)
๐•= For example v 2,11 = ร— =0.0215 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

26 TOPSIS: Example (contd..)
Calculate V+=(0.0669, , , , , , , , , , ), where ๐• + = max i v ij jโˆˆJ ,i=1,โ‹ฏ,m = v 1 + , v 2 + ,โ‹ฏ, v n + v 1 + =max ,0.0669,0.0535, =0.0669 v 2 + =max ,0.0491,0.0632, =0.0632 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

27 TOPSIS: Example (contd..)
Calculate V-=(0.0535, , , , , , , , , , ), where ๐• โˆ’ = min ๐‘– ๐‘ฃ ๐‘–๐‘— ๐‘—โˆˆ๐ฝ ,๐‘–=1,โ‹ฏ,๐‘š = ๐‘ฃ 1 โˆ’ , ๐‘ฃ 2 โˆ’ ,โ‹ฏ, ๐‘ฃ ๐‘› โˆ’ v 1 โˆ’ =min ,0.0669,0.0535, =0.0535 v 2 โˆ’ =min ,0.0491,0.0632, =0.0368 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

28 TOPSIS: Example (contd..)
Calculate the distance of each project to most positive ideal solution using ๐‘† ๐‘– + = ๐‘—=1 ๐‘› ๐‘ฃ ๐‘–๐‘— โˆ’ ๐‘ฃ ๐‘— , thus ๐‘† 1 + =0.1045 ๐‘† 2 + =0.1543 ๐‘† 3 + =0.0870 ๐‘† 4 + =0.0425 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

29 TOPSIS: Example (contd..)
Calculate the distance of each project to most negative ideal solution using ๐‘† ๐‘– โˆ’ = ๐‘—=1 ๐‘› ๐‘ฃ ๐‘–๐‘— โˆ’ ๐‘ฃ ๐‘— โˆ’ , thus ๐‘† 1 1 =0.0589 ๐‘† 2 โˆ’ =0.0477 ๐‘† 3 โˆ’ =0.0387 ๐‘† 4 โˆ’ =0.0928 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

30 TOPSIS: Example (contd..)
Calculate the relative proximity index of each alternative (which is buying the house/apartment) to the ideal solution according to formula ๐‘ ๐‘– = ๐‘† ๐‘– โˆ’ ๐‘† ๐‘– โˆ’ + ๐‘† ๐‘– + , thus ๐‘ 1 = =0.3605 ๐‘ 2 = =0.2361 ๐‘ 3 = =0.3078 ๐‘ 3 = =0.6853 TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA

31 TOPSIS: Example (contd..)
Thus the ranking is ๐‘ 4 โ‰ป ๐‘ 1 > ๐‘ 3 > ๐‘ 2 Hence Alternative # 04 is the best (position # 01) choice followed by Alternative # 01 (position # 02), then by Alternative # 03 (position # 03) and finally followed by Alternative # 02 (position # 04) TOPSIS RNSengupta,IME Dept.,IIT Kanpur,INDIA


Download ppt "IME634: Management Decision Analysis"

Similar presentations


Ads by Google