On the limited value of additional time periods in strategic facility location models
The aim of facility location planning is to locate facilities and allocate customers in an optimal way regarding the desired goal. Considering a cost minimization location model, a location planner will look for the most profitable solution of the problem. Within the planning horizon of the model, it is likely that certain parameters change over time, so applying a multi-period model instead of a static one can add value. Considering several time periods with a location and allocation decision in each period allows to find a more flexible solution to the problem. However, the effect of implicit warehousing can cause the multi-period model to return a less profitable solution than the static counterpart. Therefore, the aim of this thesis is to find a possible maximum of the value of the multi-period solution (VMPS) and to check if it always occurs. To answer the research questions, a data experiment is conducted. Within a scenario of increasing customer demand, a capacitated phase-in model is applied for instances with differing input parameters. The value of the multi-period solution for an increasing number of considered time periods is then compared. The findings of the data analysis show that no consistent maximum can be found. However, the VMPS of all instances is somehow limited. No value increases according to the number of added time periods.