Stability and Endogenous Formation of Inventory Transshipment Networks
Author(s) -
Xin Fang,
Soo-Haeng Cho
Publication year - 2014
Publication title -
operations research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.797
H-Index - 140
eISSN - 1526-5463
pISSN - 0030-364X
DOI - 10.1287/opre.2014.1324
Subject(s) - transshipment (information security) , pairwise comparison , nash equilibrium , dual (grammatical number) , stability (learning theory) , microeconomics , computer science , game theory , representation (politics) , residual , construct (python library) , business , operations research , industrial organization , economics , mathematics , computer network , artificial intelligence , art , literature , algorithm , machine learning , politics , political science , law
This paper studies a cooperative game of inventory transshipment among multiple firms. In this game, firms first make their inventory decisions independently and then decide collectively how to transship excess inventories to satisfy unmet demands. In modeling transshipment, we use networks of firms as the primitive, which offer a richer representation of relationships among firms by taking the coalitions used in all previous studies as special cases. For any given cooperative network, we construct a dual price allocation under which the network is stable for any residual demands and supplies in the sense that no firms find it more profitable to form subnetworks. Under the allocation based on the marginal contribution of each firm to its network called the MJW value, we show that various network structures such as complete, hub-spoke, and chain networks are stable only under certain conditions on residual amounts. Moreover, these conditions differ across network structures, implying that a network structure plays an important role in establishing the stability of a decentralized transshipment system. Finally, we consider the case when firms establish networks endogenously, and show that pairwise Nash stable networks underperform the corresponding networks in centralized systems.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom