Advanced Supply Chain Demand Forecasting – EyeOn

 Overview

Company Description

We are EyeOn, innovative supply chain experts. We integrate consultancy, solutions and data science to help improve forecasting and planning processes at our customers. As a graduate intern, you will be part of our data science team. Within EyeOn data science, we help customers improve their planning decisions anywhere on the spectrum from quick-wins to industry-leading advanced data science projects.

Project Description

Time series forecasting is challenging. Especially if it concerns forecasting over 100,000 time series as is the case when we do forecasting projects for most of our customers. Until recently, the most popular time series forecasting techniques were focusing on isolated time series. I.e., predicting the future of one time series considering the history of this series alone. These models are also known as “local” models. Some well-known examples include ARIMA or exponential smoothing. With the rise of Machine Learning we can also train a model on multiple (related) time series. When a model is trained on multiple time series it is referred to as a “global” model. Theoretically, global models can outperform local models since the model is able to match patterns across a collection of related time series (also known as cross learning). However to retrieve the most value from these ML approaches it is important to consider topics such as:
• Hyperparameter tuning strategies and model      architecture choices
• Choice of loss functions
• Clustering approaches (i.e., find the right cluster/collection of related series to train your model on)
• Model monitoring & MLOps
• Many more…

At EyeOn data science we are continuously working on improving our forecasting knowledge, tools and software. If you’re interested in doing research on a topic related to advanced supply chain forecasting, EyeOn might be the place for you! The exact topic will be discussed during application procedure.

Goals of project

T.b.d.

Deliverables

T.b.d.

Essential Student Knowledge

Ideally you already have (solid) programming skills and (basic) knowledge of Machine Learning. Additionally, you’re keen to apply these skills/techniques in a supply chain context.oad transport and constraints

More information: escf@tue.nl