Demand Fulfillment in Multi-Stage Customer Hierarchies

• Applicants: Professor Dr. Moritz Fleischmann; Professor Dr. Herbert Meyr; Professor Dr. Richard Pibernik
• Subject Area: Accounting and Finance
• Term: from 2014 to 2018
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 263894463

Demand fulfillment (DF) aims at optimally matching customer orders with available re-sources (i.e. capacity and inventory). In manufacturing environments, this problem is often challenging since purchasing and production need to be planned with sufficient lead time, demand is uncertain, and not all customers are of equal importance or profitability. A com-mon assumption across the state-of-the-art literature is that the overall customer base is subdivided into a single set of different customer segments, which are ordered according to their profitability/priority. Thus, a single-level allocation problem, equivalent to the perspective of a central planner, has to be solved. In reality, however, customer segments typically have a multi-level, hierarchical structure, reflecting the structure of the sales organization. Typical hierarchical levels include different geographies, different distribution channels, and different customer groups, e.g. key accounts. In such a setting, there is no omniscient central planner. Instead, information asymmetries arise. Higher-level sales managers have to make decisions based on aggregated data, rather than on detailed data of individual lower level sales organizations. In such a situation, allocation planning is an iterative process, in which higher-level sales quotas are disaggregated one level at a time by multiple local managers. For example, in a geography-based hierarchy, scarce product quantities are first allocated to continents, then to national sales organizations, and finally to regional sales districts. Thus, multiple rationing decisions are made sequentially and across multiple hierarchical levels, using locally available aggregate demand and profitability information only. In todays business practice, as reflected in state-of-the-art Advanced Planning Systems, the hierarchical DF problem is addressed by pre-defined rules, such as fixed ratios, to determine the break-down of sales quotas in a planning hierarchy. However, the available rules are simplistic; no guidelines exist as to which rule to choose in which business context and how to optimize its input parameters. Our proposed research aims to fill the existing void in research and practice and to provide scientifically sound methods for the practically relevant problem of hierarchical DF. Our goal is to develop rigorous and systematic methods for solving the hierarchical DF problem, i.e. for making medium and short-term resource allocation decisions given a multi-level customer hierarchy. The specific objectives of the proposed project are as follows: (1) Develop novel and rigorous approaches for hierarchical DF to close the aforementioned gap in scientific research and practical applications. (2) Demonstrate their feasibility, applicability, and performance, especially relative to the simplistic rules which are currently employed. (3) Derive recommendations as to when and under which conditions a certain approach should be selected.

• DFG Programme: Research Grants