Machine Learning in field demand forecasting – ASML

 Overview

Company Description

ASML is an innovation leader in the semiconductor industry. We provide chipmakers with everything they need – hardware, software, and services to mass-produce patterns on silicon through lithography.

Project Description

The Field Forecasting & Planning department at ASML is responsible for the planning of spare parts and service tools in order to meet service level agreements with ASML’s customers. The forecast of spare parts demand is an important driver for these planning activities. Our team aims to deliver an accurate field demand forecast to support the high system uptime requirements of our customers. We expect that machine learning has the potential to help achieve a higher forecast accuracy.

Trigger for project

The more accurate ASML can predict where spare part demand will occur, the more accurate they can plan to meet all the customer service level agreements. In this way, ASML can improve the material availability for their customers while decreasing the total costs. Within ASML there is a lot of information available about the installed base and parts usage that could potentially help to predict future spare part demand. ASML wants to make the next step in spare part demand forecasting by adopting the latest machine learning techniques.

Goals of the project:

A forecasting method to improve the forecast accuracy by applying machine learning techniques. This method should fit in the current framework for forecasting and planning, while making use of the available data within ASML.

Deliverables:

• Literature study to determine which methods can be used
• Conduct interviews within ASML to determine which methods are already considered and which business data is available to support these methods
• A proposal for a new ML-based forecasting method
• Model and analysis to evaluate the effectiveness of the proposed method
• Managerial guidelines for implementation

 More information: escf@tue.nl