Transparency of risk for global and complex network decisions in the automotive industry

This paper presents a multi-period, multi-objective optimization model that enables robust production network and location planning during times of increased market uncertainty and risk exposure in environmental factors. The planning model focuses primarily on integrating the quantitative and qualitative risk factors associated with strategic production network planning. Integration was realized by linking a robust programming model with aspects of multi-target optimization using an EPSILON-constraint approach. This allows quantitative risks such as fluctuations in demand, exchange rates, and transportation costs to be factored into a common mixed-integer optimization model as discrete scenarios, with qualitative risks, e.g., security risks, accounted for as part of a points-based evaluation approach. The developed model minimizes the firm's expected downside risk exposure while meeting a pre-selected threshold level for environmental risks. The trade-off between risk and profit is depicted and interpreted using a Pareto analysis. The developed model makes it possible to characterize a variety of risks that are very difficult to manage when carrying out strategic planning of production networks in practice.