The Impact of Partial Production Capacity Sharing via Production as a Service

Christina J. Liepold, Okan Arslan, Gilbert Laporte, Maximilian Schiffer

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Cloud manufacturing, one of the trends subsumed under Industry 4.0, influences industrial production settings as it transforms resource utilization into a shared on-demand service. Central to cloud manufacturing is the Production as a Service (PaaS) paradigm, which allows intermediaries to coordinate supply and demand of idle production capacities and enables collaboration in production through outsourcing. Current practical implementations of PaaS use multiple steps to connect buyers and suppliers through an intermediary that treats each demand request individually and subsequently contacts a pool of suppliers. In contrast, we study a novel optimization-based PaaS approach and develop a combinatorial exchange model for production capacity sharing among multiple suppliers and buyers. To this end, we adapt the winner determination problem from procurement auction settings. We find that key performance indicators such as matched capacity, overall payment, and unfulfilled minimum demand depend on both the employed bidding behavior and the buyer and supplier ratio. We additionally develop a corresponding online counterpart that can be readily implemented in practice and show that it yields close to optimal solutions. Finally, we show that our solution approach increases the efficiency of matching idle production capacities on average by 39% and also increases the average supplier payoff by up to 74% compared to a practice-based benchmark heuristic.

Original languageEnglish
Article number106587
JournalComputers and Operations Research
Volume165
DOIs
StatePublished - May 2024

Keywords

  • Production as a Service
  • Servitization
  • Sharing platforms
  • Winner determination problem

Fingerprint

Dive into the research topics of 'The Impact of Partial Production Capacity Sharing via Production as a Service'. Together they form a unique fingerprint.

Cite this