Abstract:Static scheduling mainly designs an available task assignment for the scheduling process according to the initial state. But it cannot meet the requirements in time and cannot response the changes agilely. As an alternative, dynamic scheduling can overcome these flaws, and it can aid to gain a more credible scheme for scheduling. As a coin has two sides, complexity occurs in the dynamic scheduling problem. Due to this, the traditional methods cannot easily build the scheduling model and cannot quickly find the approximately optimal solution or the satisfactory solution for scheduling. In this paper, a new dynamic scheduling method based on simulation of workflow is proposed, including scheduling implementation model and its matching algorithm.Firstly, the dynamic scheduling implementation model based on simulation of workflow (acronymic as DSIMBSW) is introduced. The tasks in scheduling are described in the Process Model of DSIMBSW, which is derived from the workflow model[1]. Not only can the workflow model depict the scheduling process much more easily, but also it can roundly describe many constraint conditions which exist in the scheduling. And the object function can be expressed by the statistic data which are generated through the simulation. Anyway, the constraint conditions and the object function are expressive in the DSIMBSW, as well as or even more perfect than the other models for scheduling. In this part, the components of the DSIMBSW and their relationships are detailedly explained.As the DSIMBSW involves the simulation, the algorithm based on rules can be competent. There are hundreds of rules in scheduling, and how to choose the exact rule for the exact task assignment is a problem. In order to get more satisfactory solution, dynamic scheduling algorithm based on rule-agent (acronymic as DSABRA) is proposed in this paper. The rule-agent is embedded into the Scheduler of the DSIMBSW and it has three units: Perception Unit, Analysis Unit and Action Unit. DSABRA is combined with simulation of workflow and it chooses the exact rules with the aid of rule-agent.In the last part, the scheduling implementation model based on simulation of workflow and the one based on simulation of petri-nets are compared, the merits of DSIMBSW are listed and the research of scheduling on workflow is appraised.

A framework for dynamic rescheduling problems

A multi-level rescheduling approach for a dynamic remote operations scheduling problem

A Data-Driven Simulation-Optimization Framework for Generating Priority Dispatching Rules in Dynamic Job Shop Scheduling with Uncertainties

Solving comprehensive dynamic job shop scheduling problem by using a GRASP-based approach

A Rolling Horizon Job Shop Rescheduling Strategy in the Dynamic Environment

Real-time Problem Solving Framework

Proactive Algorithms for Job Shop Scheduling with Probabilistic Durations

Dynamic job-shop scheduling using graph reinforcement learning with auxiliary strategy

A Two-stage Framework and Reinforcement Learning-based Optimization Algorithms for Complex Scheduling Problems

Deep reinforcement learning for dynamic flexible job shop scheduling problem considering variable processing times

Residual Scheduling: A New Reinforcement Learning Approach to Solving Job Shop Scheduling Problem

Multi-objective reinforcement learning framework for dynamic flexible job shop scheduling problem with uncertain events

Dynamic Job-Shop Scheduling Based on Transformer and Deep Reinforcement Learning

Dynamic Scheduling Based on Simulation of Workflow

A Deep Multi-Agent Reinforcement Learning Approach to Solve Dynamic Job Shop Scheduling Problem

Dynamic job-shop scheduling in smart manufacturing using deep reinforcement learning

A deep reinforcement learning model for dynamic job-shop scheduling problem with uncertain processing time

Deep reinforcement learning for dynamic scheduling of a flexible job shop

A MAS architecture for dynamic, realtime rescheduling and learning applied to railway transportation

A Resilient and Effective Task Scheduling Approach for Industrial Human-Robot Collaboration

Dynamic Multi-Objective Scheduling for Flexible Job Shop by Deep Reinforcement Learning.