A deep reinforcement learning based multi-criteria decision support system for optimizing textile chemical process

Textile manufacturing is a typical traditional industry involving high complexity in interconnected processes with limited capacity on the application of modern technologies. Decision-making this domain generally takes multiple criteria into consideration, which usually arouses more complexity. To address issue, present paper proposes decision support system that combines intelligent data-based models random forest (RF) and human knowledge-based multi-criteria structure analytical hierarchical process (AHP) accordance objective subjective factors textile process. More importantly, chemical described as Markov (MDP) paradigm, deep reinforcement learning scheme, Deep Q-networks (DQN), employed to optimize it. The effectiveness has been validated case study optimizing ozonation process, showing it can better master challenging decision-making tasks processes.