v Procedure Adopted in Dynamic Programming Define the variables, objective function and constraints. Divide the problem into number of sub-problems. Develop recursive relationship for optimality. Decide whether to follow the forward or the backward method to solve the problem. Make tabular presentation to show the required values and calculations for each stage. Find optimal policy at each stage and then the overall optimal policy. »Resource Allocation Problem The owner of a chain of four grocery strawberries. The table gives the estimated profits at each store when it is allocated various number of boxes: The owner does not wish to split crates between stores, but is willing to make zero allocations. Find the allocation of six crates so as to maximize the profits. No. Estimated Profits mete Sr Store sire site 0 0 0 0 0 1 4 2 6 2 2 6 4 8 i a 7 6 8 4 4 7 8 8 4 5 7 9 8 4 6 7 10 8 4Resource Allocation Problem Solution: Stage 1: Store | Stage 2: Store | + Store 2 tore 1 + Store 2 Stage Store 3 Stage 4: Store | + Store 2 + Store 3 + Store 4 Cont. Estimated Profits Store 1 Store 2 Store Store 0Resource Allocation Problem Cont... Let, = X;, XX, and x, are number of crates(boxes) allocated to store 1, store store 4 respectively. * fil), fic), FAs) and fx) are respective profit from stores. store 3 and = Maximize, Z =f; (0) + flrs) + ful) + fle) Subjected to, x +x) +x; +x, <6, Where, x), ¥3,X3,X,2 0 Stage 1: Store 1 No. of boxes, x, oki }2}3]4]sjfeo Profit, f, (x) 04) Ie) Taz areResource Allocation Problem Cont... Stage 2: Store 1+ Store 2 storey | aexrte | Od | 2 3 ]4][s]6 = 0 4 $ 7 7 7 7 Store 2 x fLodtho) Wp 0 oe Le [7 7 [7 7 1 2 2 LAN EST 9 a lol] = 2 4 waa Bieea| ires |e l 3 6 6 | se Lee Las 4 8 8 “12 Lae 9 9 6 10 | 10Resource Allocation Problem Cont... Stage 3: (Store 1 + Store 2) + Store 3 No. of boxes 0 L 2 3 4 5 6 Max. of f,(x,)+h (%)| 0 4 6 8 10 4 Boxes in 20 | pa | 22 [243] | Store1+Store2 |] 1? | adi | fiz [/a43/[ 4] 24 Stone [ies mont mile 4s [e mminal\0 [4 [6/7 1 71717 Store? x [hoy fLOdt he.) § oo lobe Let LA 7 ba Le 1 2 z wr | 87 [79 TLS 2 4 4 |_s* | s0* 7] AT TL 3 6 6" | 10°" | 42* 13 4s [8 [er le s|9 [9s «| 10 [10Resource Allocation Problem Cont... Stage 3: (Store 1 + Store 2) + Store3 No. of boxes Ce | | | |S [[Link]| o | 4 | 6 [ 8 | 10 [12] 4 Boxes in 20 22 [23 store1+Store2 | | 1 | tut iu | 1c i Max. of [f, (x,)+ f (%.)I+ fy (3) wh % ho 0 0 Ona at | 6 | SM [PLO] 120/14 1 6 A re 2 8 87 ae | a Lie [a8 3 3 8 | et | 46 4 8 8 | ae Lag Ss 8 8 A 6 8 8Resource Allocation Problem Stage 3: (Store 1 + Store 2) + Store3 No. of boxes 0 1 2 4 s 6 |Max. of f(q)+h ()| 0 4 6 10 [12 | 14 Boxes in 20 242 [243 Store iGlore 2a | Roeed |leaiead | TE 143 [144 | 24 sips Max othe hook 66) S| | hep 0 0 o* 1 4 8 10 12 14 1 6 6* 10* A2* j4* 16* | 18* Zz 8 8 h* 14* 16* 18* 3 8 8 1z At 16 4 8 8 42 44 s 8 8 AX 6 8 8 Cont.Stage 4: (Store 1 + Store2 + Store3) + Store 4 Resource Allocation Problem Cont. ty (4) No. of boxes 0 1 2 3 4[5]|[6 nee «| 2 6 10 12 | 14 | 16 | 18 + 2+0+1 eee sie re estne3 [O10 | Het | tet |g aac] 2 Boxes in Store 4 6 s 4 3 211] 0 f(x) 4 4 4 4 3 | 2] 0 fh Oa) Hs) | 4 10 14 16 | 17 | 18 | 18Resource Allocation Problem Cont... Stage 4: (Store 1 + Store2 + Store3) + Store 4 ee || Max No, of boxes (i So ed eT Max. of ee lemeltetlen optimal allocations can be as follows A (xy)+h Gy) +) num profit is 18 and possible Boxes in Store I | Store 2 | Store 3 [Store 4 oror0 | oor | 14081 Store 1+Store 2+Store3 2 2 1 1 1 3 1 1 Boresinswes | 6 | § | 4 Ls iL Boxes in 2 1 2 1 swersirersires} 4 | 4 | 4 | 4 | 3 [2] o _ 5 wees | | | | 1 2 1 Lorthad+sad | 4 | ao | uw | a6 | a7 3 1 0 HG oo AS 1 4 1 0 2 2 2 0