Results using a least-cost-per-lane solution

the resulting decision may not generate the lowest-cost decision for the supply chain. The key takeaway from this example is that myopic, single-stage optimal decisions may not generate the best result throughout the entire supply chain. But how much can the solution be improved  

A first step to improving the solution is to consider the total cost per unit along the chain from the plant through the warehouse to the customer zone. There are twelve possible chains  

Given the costs of these paths, the optimal decision for eadi customer zone would be for C1 to pick W1, C2 to pick W2, and C3 to pick W2. In turn, warehouse Wl is supplied 50,000 units by plant PI, and warehouse W2 is supplied 60,000 units by P2 and 90,000 units by PI. Given these flows, the associated cost can be verified to be 870,000, which is lower than the earlier solution. 

Note that unlike the solution in the earlier section, this approach takes account of the cost along the entire path from the plant to the warehouse to the customer zone. However, it still does not account for plant P2 s capacity when making the customer zone sourcing decision. Thus, a possible reason for the absence of a lowest-cost solution for the supply chain is that we may not have allocated plant P2 s capacity optimally across the warehouses. 

In this section, we describe the optimal solution to the problem using linear programming as a solution tool. The linear programming model takes a global look at the problem and incorporates the path of flows. 

---

Figure 2.6 Results using a least Cost-per-lane solution...

---