Data Analytics Methods for Supply Chain Risk Management with Applications in Transportation and Manufacturing Safety

Abstract

Supply chain management aims to understand and explain how organizations should collaborate within a chain to improve the overall competitiveness of the chain and smooth the flow of the money, material, and information between them. Any risks associated with each member will affect the overall performance of the chain. In this dissertation, we consider analytical modeling approaches to two specific components related to risks in supply chains, specifically the risk associated with manufacturing and transportation categories. In the first part (Chapter Two), we consider the problem of employing job rotation schemes to improve worker safety in a manufacturing setting by combining optimization methods with novel modeling techniques developed in the occupational safety community. Recent studies suggest that job rotation schedules may increase the overall risk of injury to workers included in the rotation scheme. We describe an optimization framework evaluating the effectiveness of a job rotation scheme using the fatigue failure model of MSD development and a case study with real injury data. Results suggest that the effect of job rotation is highly-dependent on the composition of the job pool, and the inclusion of jobs with higher risk results in a drastic decrease in the effectiveness of rotation for reducing overall worker risk. The study highlights that in cases when high-risk jobs are present, redesign of those high-risk tasks should be the primary focus of intervention efforts rather than job rotation. In the second part (Chapters Three, Four, and Five), the goal is to improve transportation safety in a supply chain. To do so, we first aim to reduce the start-up burden of data collection and descriptive analytics for statistical modeling and route optimization of risks associated with motor vehicles. Then, we focus on improving the safety of truck drivers. The emergence of sensor-based Internet of Things (IoT) monitoring technologies have paved the way for conducting large-scale naturalistic driving studies, where continuous kinematic driver-based data are generated, capturing crash/near-crash safety critical events (SCEs) and their precursors. However, it is unknown whether the SCEs risk can be predicted to inform driver decisions in the medium term (e.g., hours ahead) since the literature has focused on SCE predictions either for a given road segment or for automated breaking applications, i.e., immediately before the event. Here, we examine the SCE data generated from 20+ million miles-driven by 496 commercial truck drivers to address three main questions. First, whether SCEs can be predicted using disparate driving-related data sources. Second, if so, what the relative importance of the different predictors examined is. Third, whether the prediction models can be generalized to new drivers and future time periods. We show that SCEs can be predicted 30 min in advance, using machine learning techniques and dependent variables capturing the driver’s characteristics, weather conditions, and day/time categories, where an area under the curve (AUC) up to 76% can be achieved. Moreover, the predictive performance remains relatively stable when tested on new (i.e., not in the training set) drivers and a future two-month time period. Our results can inform dispatching and routing applications, and lead to the development of technological interventions to improve driver safety.