Machine Learning and LeadTime Prediction

From ISLAB/CAISR
Title Machine Learning and LeadTime Prediction
Summary Machine Learning and LeadTime Prediction
Keywords
TimeFrame Fall 2018
References
Prerequisites
Author
Supervisor Slawomir Nowaczyk
Level Master
Status Open

Generate PDF template

This is a project together with Volvo Group Truck Operations - Service Market Logistics Apply here: https://xjobs.brassring.com/TGnewUI/Search/home/HomeWithPreLoad?partnerid=25079&siteid=5171&PageType=JobDetails&jobid=663603#jobDetails=663603_5171

Background

Variations in lead time are today impacting the customer perception of service market capabilities, driving a lot of cost and tied up capital. If we could predict lead time in a better way we could give better service to our customer as well as reducing the logistics cost heavily. Since we know that lead time variations is not easy to predict with traditional time series data we believe that using other technologies, such as machine learning, could be part of the solution.

Thesis questions and expected outcome

We would like the students study current lead time setup for a number of flows and the variations and the potential of forecasting the lead time variation. Data collection, processing and quality review would be relevant during this step of the process as well. Look into technologies to forecast lead time variations such as ML, Neural networks, and others. Develop model with features, variables, and vectors. Proof of concept how the lead time forecast would benefit the supply chain.

Student profile and application

Master students in Logistics or in Mathematics, or similar fields Application deadline: Nov 5th 2018