THE INFERENCE OF HUMAN RESOURCE PLANNING OPERATION USING BAYESIAN NETWORKS

摘要

The poor state of the economies all over the world has been changing the environment of manufacturing processes and human resource planning. In particular, almost every manufacturing site has been supported by a lot of part-time, temporary, or mid-career personnel. The days when companies had a lot of talented people as regular employees and held onto them for the long term are a thing of the past. However, excellent leaders of front-line workers must design more complex human resource strategies that take into consideration the workers' skills. On the other hand, it is also a fact that these leaders must redesign daily human resource planning by hand writing, considering changeable workers' skills. Because most human resource planning systems do not consider a worker's skills, it can be effective enough to provide even the simple support tool that a sub-leader, rather than the leader, can redesign the current strategy using an interactive interface. Therefore, firstly, we propose a simple graphical interactive support tool for human resource planning using PERT concepts. The purpose is to develop the automatic graphical interactive support tool for human resource planning which has been made manually. We will be able to understand all the process by using "PERT (Program Evaluation and Review Technique) chart concept". Secondly, we discuss the effectiveness of our tool compared with redesigned time, makespan and workload. Some experiments which are evaluated by 4 steps get makespan and workload in addition to changes based on 4 cases. The result of experiments shows that the tool enabled the subject to shorten the redesigned time for human resource planning. Lastly, we make a probabilistic inference model with "Bayesian Networks" using a lot of handling records of an expert, and verify some factors to make some alternative plans easier by comparing with handling records of some non-experts.