Skip to main content

Weighted decision matrix for drive selection

Short answer

Multiplying ratings by weights gives scores: A 106, B 99, C 97, D 95. Best choice is V-belt drive (A). Limitations: subjective weights/ratings can change ranking, and linear addition may hide a very poor rating on a critical criterion.

A conveyor needs a speed-reduction drive, and four alternatives are to be evaluated: A (V-belt drive), B (chain drive), C (separate gearbox) and D (integral gear-motor). The criteria weights are: initial cost 5, reliability 4, ease of maintenance 3, compactness 2. The ratings given by the design team (scale 1 to 10, 10 best) in the order [cost, reliability, maintenance, compactness] are: A = [7, 8, 9, 6], B = [8, 6, 7, 7], C = [9, 5, 6, 7], D = [6, 8, 5, 9]. (a) Use a weighted decision matrix to rank the alternatives and state the best choice. (b) State two limitations of this method of evaluating alternatives.

A weighted decision matrix multiplies the rating of each alternative on every criterion by the weight of that criterion, and the totals are compared.

(a) Ranking

Weights: cost 5, reliability 4, maintenance 3, compactness 2 (sum = 14). Maximum possible score = 10 x 14 = 140.

Score of A =5(7)+4(8)+3(9)+2(6)=35+32+27+12=106= 5(7) + 4(8) + 3(9) + 2(6) = 35 + 32 + 27 + 12 = 106

Score of B =5(8)+4(6)+3(7)+2(7)=40+24+21+14=99= 5(8) + 4(6) + 3(7) + 2(7) = 40 + 24 + 21 + 14 = 99

Score of C =5(9)+4(5)+3(6)+2(7)=45+20+18+14=97= 5(9) + 4(5) + 3(6) + 2(7) = 45 + 20 + 18 + 14 = 97

Score of D =5(6)+4(8)+3(5)+2(9)=30+32+15+18=95= 5(6) + 4(8) + 3(5) + 2(9) = 30 + 32 + 15 + 18 = 95

AlternativeCost x5Reliab. x4Maint. x3Compact x2TotalShare of max
A V-belt353227121060.76
B Chain40242114990.71
C Gearbox45201814970.69
D Gear-motor30321518950.68

Ranking: A > B > C > D.

Answer: Alternative A (V-belt drive) is the best choice with a score of 106 out of 140.

(b) Limitations

  • The ratings and weights are subjective; a small change in weights can change the ranking (here B, C and D are within 4 points).
  • The method treats the criteria as independent and adds them linearly, so a very poor rating on one critical criterion (for example safety) can be hidden by high ratings elsewhere. Important constraints must be checked separately as pass/fail.

✓ Answer checked by Claude Opus 5.5 (high effort)

All Machine Design I chapter 1 questions →