Home Success Stories Coop Mineraloel AG: The Automation Of Rolling Monthly Projections

Coop Mineraloel AG: The Automation Of Rolling Monthly Projections

Coop Mineraloel AG Steers Safely Through The
Covid-19 Crisis Thanks To Predictive Analytics

For its monthly rolling projections, Coop Mineraloel AG now relies on a system developed and implemented successfully by s-peers AG. This system predicts sales and expenses for the entire Coop service station and Pronto store network at the click of a button – based on the latest algorithms. The advantages are obvious: Thanks to the newly gained information density and the new forecasts, logistical processes can be optimized and made more efficient.


The massive reduction in manual effort frees up capacities for new tasks while the error probability is close to zero. Thanks to the automation of the monthly extrapolation, the Controlling is able to present different possible scenarios
even in times of the Covid-19 crisis. This allows the Management to map uncertainties and to react to changes proactively.

Challenges in the Project

Coop Mineraloel AG runs a network of 314 Pronto Shops in Switzerland; 244 of these are connected, respectively, to a Coop service station. 70 of the Pronto Shops are stand-alone (e. g. in train stations). The Coop Pronto Shops – with or without a service station – are operated under a franchise system. As of the end of 2020, the Net Sales amounted to CHF 2.2 billion.

Central Controlling monitors the business development of Coop Mineraloel AG. In addition, the parent company wants to report e. g. the Key Earnings Figures on a rolling monthly basis. These requirements have so far pushed the capacities of Central Controlling to their limits: Until now, the monthly forecast was still carried out manually for each location – which, in turn, can lead to errors. For each reporting event, the large amount of data had to be merged and harmonized. Thus the reported extrapolation was very time-consuming and offered Coop Mineraloel AG only little added value in analytical terms.

In the future, it should be possible to evaluate the individual locations in the organization with a monthly extrapolation. This means that a monthly rolling sales forecast per location and differentiated by product should be available at the push of a button. By saving the previously very time-consuming manual effort, the Sales Managers thus gain more time and resources for other important tasks. By recognizing early indicators (e. g. fluctuations in Key Earnings Figures), the Controlling can also monitor developments and react to them with foresight if necessary. Furthermore, thanks to the high data density, the Sales Managers can forecast costs or other Key Earnings Figures in order to generate a monthly rolling statement about the expected annual result for their respective locations.

The implementation of the Predictive Analytics solution

By using Predictive Analytics (the automation of rolling monthly projections), Coop can increase the density of information massively. Extensive adjustments and simulations are now possible at any time on an automatically calculated database. Controlling processes are streamlined and made more efficient.

The forecasting methods used by s-peers AG are derived from classical as well as from more recent time series analyses. This means that all possible variables (e. g. seasonal patterns, trends, calendars, vacation effects) can be taken into account. Where appropriate for time series, machine learning techniques such as neural networks, classification and regression trees can also be used.

The factual granularity of the planning depends on the organizational structure as well as on the customer requirements. It differs strongly between the sectors of Industry, Service and Trade. For the temporal granularity, the calendar month is usually defined. The time horizon of the planning (e. g. how many future periods are to be forecast) can be fixed or flexible. For example, the horizon in rolling planning becomes shorter and shorter with each new forecast. Simultaneously, the oldest periods that were planned/predicted originally have
already been replaced by the actual data.

The convincing project result is a clearly structured flow of the monthly forecast regarding the sales of the individual locations: A total of twelve monthly annual sales forecasts (rolling and automatically generated) on site level is shown – as well as sales differentiated by fuel type and store sales. Risks of possible error sources and coordination breakdowns have been largely minimized; regular reporting runs uniformly and smoothly.

By using the forecasts, also Planning events can now be designed more efficiently to generate suggested values which are then transferred to the Budgeting process. All of these forecast values are created in a fully automated way. On this basis, the Planning Managers can modify the figures if a drop in sales is to be expected due to unplanned or uncontrollable circumstances – e. g. in consequence of a construction site at a certain location.

Achieved Result

Long-term rolling forecasts can now be generated on a monthly basis while long-term weekly or daily forecasts are also available. In addition, costs per site can also be predicted; this enables early and long-term site assessment. In principle, complete forecasts for all 314 locations are possible now. In addition, the procedures can also be applied at company level (e. g. for long-term forecasts of heating oil sales volumes and turnover).

At the push of a button, Central Controlling benefits from high analytical added value in the context of short-, medium- and long-term site assessment – but also for evaluating Coop Mineraloel AG as a whole. New information about locations is generated. At the same time, the cooperation processes between Central Controlling and the Sales Managers can be designed more efficiently. On the one hand, automation eliminates many manual activities; on the other, intelligent analyses help to support the monitoring of data quality. Thanks to the newly gained information density and the new forecasts (e. g. regarding the sales volumes of individual products), logistical processes can be optimized as well. In addition, the freed-up time capacities allow the Management to carry out unscheduled planning events spontaneously.

Benefits

  • Less effort for activities with no added value
  • More precise results and greater transparency thanks to increased information density
  • Clarification of interrelationships and cross-functional dependencies (across a wide range of business areas) by disclosed cause-and-effect chains
  • Massively increased responsiveness through automation – calendar orientation increasingly becomes a thing of the past
  • Objectified forecasts providing a fact-based view
  • Multiple use and reusability of algorithms provided by s-peers AG
  • Wide range of evaluation options in assured quality
  • Early possibility of performance evaluation through simulation

Verwendete Technologien

 

Published by:

Dr. Stefan Lieder

ehem. Leiter Data Science Werkstatt

AUTOR:IN

Wie hat der Artikel gefallen?

Wie hilfreich war dieser Beitrag?

Klicke auf die Sterne um zu bewerten!

Durchschnittliche Bewertung 5 / 5.
Anzahl Bewertungen: 1

Bisher keine Bewertungen! Sei der Erste, der diesen Beitrag bewertet.

Share on twitter
Share on facebook
Share on xing
Share on linkedin
Share on email

Andere interessante Artikel