Date of Graduation


Document Type


Degree Name

Bachelor of Science in Industrial Engineering

Degree Level



Industrial Engineering


Rossetti, Manuel

Committee Member/Reader

Zhang, Shengfan


This research presents a methodology for modeling the amount of demand and the time intervals between demands within a complex supply chain simulation program. The research outlines the approach, incorporating factor analysis and consolidation techniques, with a focus on using Random Forest modeling for factor importance identification. The results section covers demand quantity modeling utilizing both continuous and discrete empirical approaches, accompanied by considerations for outlier and overall distributions. While using these modeling techniques, P-P plots and KS-Statistics were employed to assess the goodness of fit, as a way to make recommendations for these models. The time between demand modeling is explored, also encompassing continuous and mixture approaches, along with outlier and overall distribution modeling. Additionally, the research investigates the non-stationary compound Poisson process behavior these variables appear to exhibit. This holistic approach provides insights into some of the factors influencing demand and how to model it for supply chain simulations.

An analysis of the time between demand process indicated that the demand occurrence process should be modeled by considering the time varying nature of the underlying stochastic process. For simplicity and based on an analysis of the time between demand process, a non-stationary Poisson process is recommended. In addition, the demand amount representing the quantity of demand when a demand occurs should be modeled with an empirical probability distribution. This results in representing the demand process with a compound non-stationary Poisson process.


Supply Chain Simulation, Random Forest Modeling, Non-Stationary Poisson Process, Distribution Fitting, Factor Analysis, Goodness-of-Fit


This study adeptly applies advanced statistical techniques to enhance the accuracy of demand modeling in supply chain simulations.

Available for download on Saturday, April 25, 2026