Production and lead time quotation under imperfect shop floor information

Abstract

We study the value of shop floor information for a manufacturer that operates under a make-to-stock queue scheme and quotes lead times to arriving customers. Two settings are analyzed where in the first setting the manufacturer is able to perfectly track the shop floor status, and in the second setting the manufacturer makes decisions under imperfect information. Under imperfect information, the problem is modeled as a Partially Observable Markov Decision Process. In the computational analysis, first the impact of information distortion on the policies is analyzed and it is shown that the increased likeliness of observing the true status of the system does not necessarily lead to higher profit. Further numerical analysis shows that as the inaccuracy in the data (range of error) increases, the value of information increases. The value of information is likely to be higher under high holding cost and low traffic. Then, the effect of Make-to-Order (MTO) and Make-to-Stock (MTS) queue schemes, as well as the effect of different lead time quotation schemes on the value of information are evaluated. Under MTO queue scheme shop floor information is more valuable under high traffic intensity, whereas under MTS queue scheme information is more valuable under lower intensity. Finally, the analysis reveals that if a firm quotes more precise lead times, then the need for actual shop floor information increases.