Research On Two-sided Matching With Matching Mechanism Design For Cloud Manufacturing Resource Under The Internet Of Things

The concept of the Internet of Things,IoT,has been widely investigated in business and academia since the day it was proposed.On one hand,innovation in production and operations management based on the IoT technology,changes the way of manufacturing and trading among global companies.On the other hand,sharing economy,which comes along with the development of IoT technology,is reshaping the public’s way of consuming and living.In the field of manufacturing,enterprises are faced with the conflict of a lack or surplus of resources in a fast changing market,while the Io T form the technological basis for information and resource sharing among them.Cloud manufacturing and other advanced manufacturing come into being in this way.On a cloud manufacturing platform,companies with surplus resources may sell them and become suppliers,while those short of resources want to purchase the ones in sale as demanders.Each supplier must be matched to one or some of the demanders before the deal is made,which is exactly what this paper focus on.Step one,a systematic analysis of manufacturing resource allocation under the IoT is made,and we just consider the one-to-many matching situation here.At first,we introduced three manufacturing scenarios at different levels of decentralization under the IoT,which are totally decentralized,monopolistic and organizational.Then the entire matching process was described in order,and three customized algorithms,namely,G-S,TTC,and Munkres are proposed,which deal with resource allocation in decentralized manufacturing environment.At last,these algorithms were assessed and compared with respect to desired properties of matching mechanisms such as stability,consistency,Pareto efficiency and so on.According to the analysis,we conclude that G-S adapts to totally decentralized scenario,that Munkres adapts to organizational scenario and that TTC adapts to organizational scenario.Step two,considering the quantity of each deal,we design a matching mechanism with agents’ expectation for cloud manufacturing resource.With multi-attribute evaluation under prospect theory,we calculated satisfaction degree values and corresponding preferences of both sides.And based on the preference information of each agent,we designed a heuristic algorithm to obtain a stable matching result for the many-to-many matching problem with matching quantities.We demonstrated that the matching scheme obtained was a Pareto-optimal solution for agents on a dominating side.A sample application finally verified the feasibility and validity of the whole matching mechanism.