← 2020 Paper 2
UPSC 2020 Maths Optional Paper 2 Q4c — Step-by-Step Solution
20 marks · Section A
Transportation problem · Linear Programming · asked 5× in 13 yrs · Read the full method →
Question
Find the initial basic feasible solution of the following transportation problem by Vogel’s approximation method (VAM) and use it to find the optimal solution and the transportation cost. Cost matrix (rows = sources S1,S2,S3; cols = destinations D1,D2,D3,D4):
S1: [10, 0, 20, 11], availability 15
S2: [12, 8, 9, 20], availability 25
S3: [0, 14, 16, 18], availability 10
Demand: D1=5, D2=20, D3=15, D4=10.
Technique
Vogel’s Approximation Method for the initial BFS (penalty = gap between two least costs), then MODI/u–v method for optimality; one loop pivot (θ=5) reaches the optimum.
Solution
| D1 | D2 | D3 | D4 | Supply |
|---|
| S1 | 10 | 0 | 20 | 11 | 15 |
| S2 | 12 | 8 | 9 | 20 | 25 |
| S3 | 0 | 14 | 16 | 18 | 10 |
| Demand | 5 | 20 | 15 | 10 | 50/50 |
Step 0 — Check balance
Total supply =15+25+10=50; total demand =5+20+15+10=50. Balanced, so no dummy needed. A basic feasible solution will have m+n−1=3+4−1=6 occupied cells.
Step 1 — Vogel’s Approximation Method (penalties = difference of two smallest costs)
We repeatedly compute, for each remaining row/column, the penalty = (2nd smallest cost) − (smallest cost), pick the row/column with the largest penalty, and allocate the maximum to its least-cost cell.
Allocation 1. Penalties — Rows: S1 ∣0,10∣→10, S2 ∣8,9∣→1, S3 ∣0,14∣→14; Cols: D1 ∣0,10∣→10, D2 ∣0,8∣→8, D3 ∣9,16∣→7, D4 ∣11,18∣→7. Largest =14 (S3). Least cost in S3 is cell (S3,D1)=0. Allocate min(10,5)=5 → x31=5. D1 satisfied (remove column D1); S3 left with 5.
Allocation 2. Remaining cols D2,D3,D4. Penalties — Rows: S1 ∣0,11∣→11, S2 ∣8,9∣→1, S3 ∣14,16∣→2; Cols: D2 ∣0,8∣→8, D3 ∣9,16∣→7, D4 ∣11,18∣→7. Largest =11 (S1). Least cost in S1 (over D2,D3,D4) is (S1,D2)=0. Allocate min(15,20)=15 → x12=15. S1 exhausted (remove row S1); D2 left with 5.
Allocation 3. Rows S2,S3; cols D2,D3,D4. Penalties — Rows: S2 ∣8,9∣→1, S3 ∣14,16∣→2; Cols: D2 ∣8,14∣→6, D3 ∣9,16∣→7, D4 ∣18,20∣→2. Largest =7 (D3). Least cost in D3 is (S2,D3)=9. Allocate min(25,15)=15 → x23=15. D3 satisfied (remove D3); S2 left with 10.
Allocation 4. Rows S2,S3; cols D2,D4. Penalties — Rows: S2 ∣8,20∣→12, S3 ∣14,18∣→4; Cols: D2 ∣8,14∣→6, D4 ∣18,20∣→2. Largest =12 (S2). Least cost in S2 (over D2,D4) is (S2,D2)=8. Allocate min(10,5)=5 → x22=5. D2 satisfied (remove D2); S2 left with 5.
Allocation 5 & 6. Only D4 remains (demand 10) with S2 (5) and S3 (5). Allocate x24=5 and x34=5. All supplies/demands met.
Initial BFS (VAM)
| D1 | D2 | D3 | D4 |
|---|
| S1 | – | 15 | – | – |
| S2 | – | 5 | 15 | 5 |
| S3 | 5 | – | – | 5 |
Occupied cells =6=m+n−1 (non-degenerate). Cost:
ZVAM=0⋅15+8⋅5+9⋅15+20⋅5+0⋅5+18⋅5=0+40+135+100+0+90=365.
Step 2 — Optimality test by MODI (u–v method)
Set u1=0 and solve ui+vj=cij on the 6 basic cells {(1,2),(2,2),(2,3),(2,4),(3,1),(3,4)}:
u1=0, u2=8, u3=6;v1=−6, v2=0, v3=1, v4=12.
Opportunity costs Δij=cij−ui−vj for the non-basic cells:
| cell | Δ | cell | Δ |
|---|
| (1,1) | 16 | (1,3) | 19 |
| (1,4) | −1 | (2,1) | 10 |
| (3,2) | 8 | (3,3) | 9 |
Δ14=−1<0, so the solution is not optimal; cell (S1,D4) should enter the basis.
Step 3 — Loop pivot on entering cell (S1,D4)
Closed loop with alternating +/− signs starting at (1,4):
(1,4)+→(2,4)−→(2,2)+→(1,2)−→(1,4).
The minimum allocation on the "−" cells is θ=min{x24,x12}=min{5,15}=5. Adjust:
x14=0+5=5,x24=5−5=0,x22=5+5=10,x12=15−5=10.
(2,4) leaves the basis.
New solution
| D1 | D2 | D3 | D4 |
|---|
| S1 | – | 10 | – | 5 |
| S2 | – | 10 | 15 | – |
| S3 | 5 | – | – | 5 |
Step 4 — Re-test optimality (MODI)
New potentials (u1=0): u2=8, u3=7; v1=−7, v2=0, v3=1, v4=11. Opportunity costs for all non-basic cells:
Δ11=17, Δ13=19, Δ21=11, Δ24=1, Δ32=7, Δ33=8.
All Δij≥0 ⟹ optimal.
Step 5 — Optimal cost
Z\*=0⋅10+11⋅5+8⋅10+9⋅15+0⋅5+18⋅5=0+55+80+135+0+90=360.
Answer
Optimal: x12=10, x14=5, x22=10, x23=15, x31=5, x34=5;Z\*=360.