Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
This paper presents a methodology for eliciting the behavioural model of purchasers' decision-making in a supply chain environment. The objective of the work is to explicitly describe the relationship between a human user's behaviour and the resulting performance in inventory level and service satisfaction. The key research findings include: (i) How to identify a categorised pattern of a human decision model, particularly concerning purchasing operations; and (ii) Which metrics would be relevant in the objective of carrying out a quantitative analysis on decision behaviour. The order placement behaviour is studied as one of daily operational decisions. The work demonstrates how to analyse a potential relationship between each category of behaviour and selected measures of supply chain performance, i.e., average inventory level and delivery satisfaction. We developed a specialised tool for supply chain simulation aimed at collecting data such as information consultations and decision-making actions of purchasing operations. The methodology is composed of sub-processes, such as data log file generation, data parsing and information generation, behaviour profile identification through clustering, and the analysis of supply performance regarding each category of decision-making behaviour. Through experimentation with industrial purchasing agents, we validated the effectiveness of the approach and demonstrated how to achieve a quantitative analysis of decision making behaviour and its impact on supply performance. (C) 2017 Elsevier Ltd. All rights reserved.