Staff scheduling for Fiege Austria
Forecasts that are created on the basis of Excel have the major disadvantage that they are based purely on historical, internal data. Our pacemaker Forecast tool enabled Fiege Logistik to plan staffing levels for a customer in the electronics retail sector based on internal data and automatically adding important, external factors.
The Fiege Logistik branch in Vienna employs just under 110 people on a warehouse area of 16,000m². Fiege handles all logistics operations for its customers. The customers also include a large electronics retail chain. A customer that cannot provide forecasts and also has a strongly action-driven business model. Accurate forecasts of incoming and outgoing goods quantities in the B2B and B2C business are necessary here in order to be able to implement long-term, data-based workforce planning.
For an electronics retail chain, Fiege Logistik previously had internal forecasts that were used on an Excel basis for staff scheduling. With this, it was only possible to react manually to events and peaks. Long-term vacation planning and targeted staff expansion are thus hardly feasible. Increased costs are incurred and the forecasts are based on purely internal data without taking into account important external factors such as major sporting events, the effects of the COVID-19 pandemic and other influences on incoming and outgoing goods.
At the Vienna site, the AI-based forecasting tool pacemaker Forecast was made available for Fiege Logistik. With the fully automated SaaS solution, the site was able to predict incoming and outgoing goods for the next few weeks. As a result, staff could be planned far more long-term. For the first time, external influencing factors were also included in the forecasts of incoming and outgoing goods. For example, the tool showed that special effects such as major sporting events and the corona-related closure of stores had a particular impact on the site’s capacity utilization.
Together with Michael Jahn (Managing Director Fiege Austria) and the team of the Fiege Omnichannel Retail site, a fully automated Machine Learning Forecast could be implemented into the workforce planning within a few weeks. Already the first forecast achieved an accuracy of over 90% and was improved with further data reconciliations. In personnel requirements planning alone, Fiege Logistik was thus able to record annual cost savings of around €10,000.