Predicting Human Decision-Making for Task Selection in Manufacturing: A Systematic Literature Review

Predicting human decisions is a central challenge for planning and controlling production with weakly structured processes. Thus, workers' decisions regarding the processing strategies and the temporal sequence of tasks to be processed are to be determined prospectively. Accordingly, there is a need to review methods for preference elicitation to develop individual predictive decision models. This paper presents a systematic literature review and discussion of 42 publications on predictive decision models and decision attributes. Methods for eliciting decision-making knowledge from manufacturing workers as part of the modeling process and decision model validation methods are reviewed and discussed in light of their predictive validity for individual task selection. The article synthesizes the recent literature for predicting human decision-making in manufacturing using artificial intelligence methods. Along with the review results, a future research agenda is proposed for modeling and simulating human decision-making in manufacturing. Knowledge about human preferences and the successful prediction of workers' decision-making in manufacturing helps companies predict manufacturing objectives and derive organizational and work design measures.