Pricing models with AI for the Next Generation (PANG)
| Reference number | |
| Coordinator | Proj.inriktad Forskning O Utveckling i Götebor - Proj Inriktad Forskning & Utveckling I Göteborg |
| Funding from Vinnova | SEK 417 600 |
| Project duration | May 2020 - March 2021 |
| Status | Completed |
| Venture | AI - Competence, ability and application |
| Call | Start your AI-journey! For businesses |
Important results from the project
The purpose of the project was to contribute to a more energy-efficient, sustainable and flexible Swedish district heating through a broadening of AI-related competence. The objective was to develop methods and tools that can contribute to creating improved and understandable pricing models that reward active energy and power efficiency in district heating networks. More specifically, the project focused on further developing data-driven methodologies to: - Evaluate AI-supported analysis - Suggest dynamic pricing models - Predict heating load - Test ´clustering´ algorithms
Expected long term effects
In this project, we have identified that there are a lot of opportunities with using digitalization and AI in the analysis of pricing models for the district heating industry. Using clustering algorithms has sparked interesting discussions and together with algorithms for predicting heat demand, we have been able to propose two new pricing models for district heating based on dynamic pricing - ie the price is determined dynamically (for example) a day before. We hope to apply these pricing models in a real scenario as a spin-off project in the near future.
Approach and implementation
May 2020: Start-up and first review of available data. Introduction of the project schedule to external partners. June-Sep 2020: Data processing and knowledge transfer on machine learning Sep-Nov 2020: Model development Dec 2020: Testing and validation, delivery of preliminary results to external partnerships