Abstract
Supply chain system (SCS) is network control system composed of several subchains, each subchain in SCS has computation intelligence and information interaction capability could be described as an agent. Such that the SCS could be modeled as multiagent system. Investigating H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> leader-follower consensus problem could solve the product quantity matching problem and attenuate the bullwhip effect caused by uncertain market demand, which is considerable to improve system performance for SCS. In the framework of zero-sum graphical game, production rate and uncertain market demand as game players form a confrontation relationship. The optimal production rate and worst-case uncertain market demand related to zero-sum graphical game solution could be obtained at Nash equilibrium. Due to acquire game solution depends on solving coupled Hamilton-Jacobi-Isaacs (HJI) equation, the value iteration algorithm is introduced and the actor-critic-disturbance neural network structure is presented for approximating game solution, optimal production rate and worst-case uncertain market demand. Lastly, a numerical simulation and result analysis are provided to prove the effectiveness of adopted method.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call