Global Production Networks – An Approach to Find the Optimal Operating Point in the Conflict Between Risk- and Cost-minimization

AbstractThe proceeding globalization leads to larger and more complex production networks. Especially the production industry forces companies to distribute their production sites around the globe by high cost pressure. Low labor costs, transport costs, distances to potential markets and productivity influences lead to complex decision situations for managers in charge. Besides the total landed costs to produce the product, portfolio decisions are influenced by external risks, causing fundamental losses like recent past shows. This makes decisions even more complex and leads to the conflict of risk- and cost- minimization. Therefore, an integrated methodology is needed to determine risks and costs of a global production networks. Furthermore, a procedure has to be developed to find the optimal operating point in the area of tension between risk- and cost-minimization. The approach contains the development of a quantitative mathematical risk evaluation for global production networks including differentiated dimension probabilities of occurrence and extent of loss and their combination. For cost calculation the web tool “OptiWo” is used, providing the needed detail level of the production network configuration. Combining the area of tension of risk and cost, the coherence between these two dimensions of global footprint design decisions can be identified. Furthermore, the risk optimized network structure for each level of total landed costs can be identified. Depending on the utility function concerning the risk aversion of the manager in charge or company, an individual network operating point should be detected. To make this approach effectively usable for decision makers, an intuitive and progressively resolving visualization will be developed to obtain a deep understanding of the decision situation. The methodology is supposed to be integrated in a web tool to be adaptable efficiently to every use case. The results of the approach will be validated by industry based use cases