Managing service-level contracts in sales hierarchies

Service-level contracts are increasingly common in the B2B relationships of manufacturers and their customers. Service-level contracts allow a certain part of demand to remain unfilled without incurring penalties, thus making allocation planning particularly difficult because the value of an allocation becomes clear only after all demands have realized. Oftentimes the hierarchical structure of sales organizations adds another level of complexity: Instead of having a central planner determining the allocations to all customers simultaneously, allocation planning must be performed decentrally along the individual levels of the sales hierarchy. We show that the resulting allocation problem is composed of two hierarchical subproblems, where the base-level problem is that of allocating supply to individual customers (customer-allocation problem—CAP), and the top-level problem is that of allocating along the individual levels of the hierarchy (hierarchy-allocation problem—HAP). We propose several new allocation approaches for the HAP and analyze their performance in an extensive numerical experiment. Our analyses suggest that with appropriate allocation approaches the performance of decentral planning can be close to that of central planning. Based on our results we also provide a useful guideline to decision makers to choose “appropriate” allocation approaches based on structural properties of the sales hierarchy.