Affiliation:
1. Integrated Systems Engineering, The Ohio State University
2. Industrial and Systems Engineering, IIT Kharagpur
3. School of Health and Rehabilitation Sciences, College of Medicine, The Ohio State UniversityCollege of Medicine, The Ohio State University
Abstract
There is a great need for creating schedules that are optimized. Yet, some individuals have had less than desirable experiences with “optimal” scheduling. This could have been due to prioritization of the wrong criteria, leading to schedules that did not make practical sense, or that were math-intensive and were not able to be easily interpreted. Also, there are many types of optimization problem formulations and solution methods. Here, we divide the formulations into two major types: batched and online scheduling classes are discussed. A different technique has been created that allows schedules to be made that are not only optimal, based on the formulations or framing, but that are actually useful. Here, we discuss two types of methods, one batched called Genetic Algorithms with an Earliest Due Date encoding Method (GAGEDD) and the other online called Markov Decision Processes and Reinforcement Learning extensions. These methods are already being employed to create practical and optimal schedules that can include many different constraints and are able to instantly take into account new scheduling requests and take optimal actions regardless of what state the system is currently in. Especially with current world events (COVID-19), it is important to intelligently schedule patients.
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献