Verification of the market potential for new forecasting method in the banking and industrial sectors.
Reference number | |
Coordinator | F&A forecasting AB - Agila labbet |
Funding from Vinnova | SEK 500 000 |
Project duration | August 2015 - July 2016 |
Status | Completed |
Purpose and goal
The aim of the project was to develop a prototype of a user-friendly forecasting-platform, where advanced statistical models could be automated and applied by non-statisticians. Included was also a module to manage qualitative information, such as forecast adjustments, by applying a machine learning algorithm. The objective was to evaluate the customer´s interaction with the tool and to compare forecast accuracy before and after the use of the platform.
Expected results and effects
As the development of the platform took longer than expected, forecast accuracy could not be evaluated since this required that users applied the tool for at least 6 months. The customers´ interaction could, however, be evaluated with very positive results. Through the Vinnova project, we have managed to establish a proof of concept for the platform. Due to the functioning prototype, we have now got the first paying customer. This, in turn, has led to interests from several investors from whom we have received several investment offers.
Planned approach and implementation
The prototype was delayed due to the team initially consisted of senior developers who were unfamiliar with the new technologies required by the platform. This delay, combined with problems related to the technical solution, led to the prototype not being in place until March 2016. After clearing the system from bugs, the platform was tested and demonstrated for the customers during the spring. If we were to redo the project all over again, we would have put less emphasis on recruiting senior developers and more focus on finding ambitious junior developers with updated knowledge.