Customer Acceptance of a Revenue Management Platform with Multi-Party Computation

Abstract

Research ProblemIn today’s competitive and fast-paced nature of conducting business in the semiconductor industry, the discipline of revenue management (RM) is often mentioned. Through a dynamic pricing capability, RM enables firms to maximize profits by capitalizing on missed revenues. Moreover, RM enables customers to receive products on short notice, by allowing them to place orders with a(n) (earlier) delivery wish date, wish price and volume. Since the semiconductor industry is characterized by volatile market demand, long production times and high capital investment (Ehm & Ponsignon, 2012), the semiconductor industry is a model representation of the manufacturing industry. Despite the benefits of RM, the manufacturing industry has not yet witnessed a widely-adopted revenue management system.In order to apply revenue management in the manufacturing industry, it is primarily important to understand that the data involved from the parties participating in the revenue management process is confidential. For manufacturers, the confidential data includes the daily unallocated available-to-promise data, which is available to commit to customers for orders on short notice. On the other hand, for customers, the maximum price that they are willing to pay for such products with a shorter delivery time is confidential. Hence, due to the confidentiality involved, as well as the need for enabling a more equal collaboration model between buyers and sellers, a secure technology that involves inputs from multiple parties is required in order to facilitate the implementation of revenue management in the manufacturing industry.Multi-party computation is a promising cryptography technology which has been theoretically studied by researchers but has not yet been practically applied. This technology is a value-added tool in the use case of revenue management because it deals with multiple parties’ confidential data and performs a computation with them securely. The outcome of the algorithm presents a revised price (based on the desired price adder range provided by the manufacturer, as well as the price wishes of the customer), a volume and a delivery date. This will allow manufacturers to secure their unallocated available-to-promise data, and customers to share the confidential maximum price they are willing to pay to receive products earlier. Multi-party computation has been proposed to be applied to revenue management by the EU H2020 SafeDEED project. We further build on this concept (SafeDEED, 2018).As a consequence of multi-party computation being an up-and-coming technology, as well as revenue management being new to the business-to-business field, it is challenging to imagine what a possible implementation of their integration is...