*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Predictive Demand Response Modeling for Logistic Systems Innovation and Optimization
Date: 5th Dec 2022
Time: 10 am – 12 noon
The meeting link: Click here to join the meeting
Zeynab Bahrami Bidoni
Machine Learning Ph.D. candidate
H. Milton Stewart School of Industrial and Systems Engineering
Georgia Institute of Technology
Committee Members:
1 Dr. Benoit Montreuil (Advisor), ISYE, Georgia Institute of Technology
2 Dr. Kamran Paynabar, ISYE, Georgia Institute of Technology
3 Dr. Yao Xie, ISYE, Georgia Institute of Technology
4 Dr. Frederick Benaben, ISYE, Georgia Institute of Technology
5 Dr. Nico SCHMID, IÉSEG - School of Management
Abstract
Purpose In the ever-increasing dynamics of global business markets, logistic systems must optimize the usage of all possible sources to innovate continually. Scenario-based demand prediction plays an important role in the effective economic operations and planning of logistics. However, many uncertainties and demand variability, which are associated with innovative changes, complicate demand forecasting and expose system operators to the risk of failing to meet demand.
This study aims to present a new approach to predictively explore how customer preferences will change and consequently demand would respond to the new setup of services caused by an innovative transformation of the logistic layout. The critical challenge is that the demand (customer) responses to the innovative changes and corresponding adjustments are uncertain and unknown in practice, and there is no historical data to learn from and directly support the predictive model.
In this dissertation, we are dealing with three different cases of predictive demand response modeling which have been presented in the following chapters. Chapter 1 provided a novel approach for predictively modeling probabilistic customer behavior over new service offers which are much faster than ever done before by a large Chinese parcel-delivery service provider. In Chapter 2, an Interactive risk analysis tool has been proposed for predicting scenario-based erection-site demand schedules under uncertainty of disruptive events in construction projects whose logistics transformed from traditional to modular style. To advance in their logistics designs and capacity adjustments, and also to enhance their capability for taking more market share, it is crucial to estimate potential future demand for modular construction and corresponding probable projects in terms of their potential location, size, and characteristics. For this purpose, Chapter 3 introduces a methodology to estimate scenario-based future potential demand (projects) for modular construction with implementation over the US metropolitan statistical areas.
