Measurement of operational complexity of supplier-customer system using entropy - a case study

The paper concerns with analysis of operational complexity of company supplier-customer relations. Well-known approach for measuring that operational complexity is based upon entropy. However, there are several approaches thereon. In the first part, we discuss various general measures of uncertainty of states, the power entropies in particular. In the second part, we use Shannon entropy as a base framework for the case study - supplier system of the most important commodity in brewery industry, the malted barley. We assume a problem-oriented database exists, which contains detailed records of all product orders, deliveries and forecasts both in quantity and time being scheduled and realized, within time period given. We concern ourselves with quantity perturbations since they play more crucial role as the time variations. The general procedure elaborated consists of three basic steps - preprocessing of data with consistency checks in Java, calculation of histograms and empirical distribution functions, and finally, evaluation of conditional entropy. These two last steps are realized by Mathematica modules. Illustrative results of operational complexity of malted barley supplier system focused on volume quantity variations are presented in detail.