Headquarter-centered Common Order Management: a simulation approach

Abstract

This paper considers the customer order management in a group corporation which consists of one headquarter (HQ) and several geographically dispersed and operationally semi-autonomous production subsidiaries. Two order management models are formulated. One is the Headquarter-centered Common Order Management (HQ-COM) where customer orders are processed by the HQ and then allocated to its subsidiaries. The other is Subsidiary-Autonomous Order Management (SD-AOM) where subsidiaries take and process customer orders relatively independent of each other. The two models are applied to study two situations, respectively. One situation is that the total quantity of customer orders exceeds the production capacity of each individual subsidiary so that the order has to be split before allocating amongst multiple subsidiaries. The other situation is that the quantity of selected customer orders is within the production capacity of a single subsidiary so that orders should be merged into one batch before allocating to one subsidiary. Different heuristics are proposed and a series of simulation experiments are conducted. The results show that HQ-COM outperforms SD-AOM in terms of both its performance and its robustness against the demand variability. This achievement is largely due to the effects of pooling of different customer orders and sharing of production capacity amongst subsidiaries.