Get the opportunity to tackle a real-life problem that brick & mortar retail faces and work in cooperation with Markant, a large German retail technology supplier, and find out, in which categories, AI-based self-learning models can yield reliable predictions for future consumers demand.
In terms of purchasing behavior analysis, brick-and-mortar retail still has fundamental disadvantages compared to online retail, where the tools for data collection and analysis are sophisticated and precise.
Yet, those tools are of high importance for detecting trends and demand patterns in order to plan and reorder inventory, or for identifying and implementing suitable (advertising) measures. In bricks-and-mortar retail the amount of data available is much smaller, or the data that can be extracted is not yet integrated into corresponding analyses to carry out optimizations in the typical fields of activity of bricks-and- mortar retail to increase sales, measure customer frequency and margins.
In particular, demand forecasting using AI and datasets such as historic transaction data, weather data and event data could help to avoid repetitive manual reordering tasks, excessive or out of stock inventories, or even food waste.
In terms of purchasing behavior analysis, brick-and-mortar retail still has fundamental disadvantages compared to online retail, where the tools for data collection and analysis are sophisticated and precise. Yet, those tools are of high importance for detecting trends and demand patterns in order to plan and reorder inventory, or for identifying and implementing suitable (advertising) measures. In bricks-and-mortar retail the amount of data available is much smaller, or the data that can be extracted is not yet integrated into corresponding analyses to carry out optimizations in the typical fields of activity of bricks-and- mortar retail to increase sales, measure customer frequency and margins. In particular, demand forecasting using AI and datasets such as historic transaction data, weather data and event data could help to avoid repetitive manual reordering tasks, excessive or out of stock inventories, or even food waste.
In collaboration with Markant, a large German retail technology supplier, you will find out, in which categories, AI-based self-learning models can yield reliable predicitions for future consumer demand. You will have access to transaction records, product information, price and promotion activities.
In collaboration with Markant, a large German retail technology supplier, you will find out, in which categories, AI-based self-learning models can yield reliable predicitions for future consumer demand. You will have access to transaction records, product information, price and promotion activities.
The thesis is hosted by:
ETH AI Center Universitätsstrasse 6, CAB E 72 8092 Zürich
klaus.fuchs@ai.ethz.ch
and issued in cooperation with:
Markant Handels- und Industriewaren AG Churerstrasse 166
8808 Pfäffikon SZ
info@markant.com
The thesis is hosted by: ETH AI Center Universitätsstrasse 6, CAB E 72 8092 Zürich klaus.fuchs@ai.ethz.ch
and issued in cooperation with: Markant Handels- und Industriewaren AG Churerstrasse 166 8808 Pfäffikon SZ info@markant.com