Periodic Reporting for period 1 - PACE (Packaging with AI for Circular Economy - strengthening Digimind's market position)
Período documentado: 2022-06-01 hasta 2023-01-31
Digimind GmbH is building software tools to enable the more than 35,000 packaging designers and engineers worldwide to tackle the technical challenges they are facing when redesigning or optimizing packaging products for circularity, thereby supporting the EU goals set in the circular economy action plan.
Digimind is researching and developing a SaaS (Software as a Service) solution using machine learning models to help packaging engineers quickly lightweight the package, increase the amount of recycled material and test the design for both performance and environmental impact. We develop software features that are tailored to address the three pillars of circular economy – Reduce, Recycle & Reuse - in near real time thanks to the unique accelerated simulations using machine learning.
This project has been used to achieve two key points to strengthen's Digimind position in the market with additional proof point of the transferability of our technology to different packaging categories.
The first activity focused on building a use case for glass packaging with an experimental testbed.
The second part focused on strengthening the robustness of Digimind’s business plan by enhancing the financial projections, provide an update of the market research and built an updated pitch deck.
Digimind is researching and developing a SaaS (Software as a Service) solution using machine learning models to help packaging engineers quickly lightweight the package, increase the amount of recycled material and test the design for both performance and environmental impact. We develop software features that are tailored to address the three pillars of circular economy – Reduce, Recycle & Reuse - in near real time thanks to the unique accelerated simulations using machine learning.
This project has been used to achieve two key points to strengthen's Digimind position in the market with additional proof point of the transferability of our technology to different packaging categories.
The first activity focused on building a use case for glass packaging with an experimental testbed.
The second part focused on strengthening the robustness of Digimind’s business plan by enhancing the financial projections, provide an update of the market research and built an updated pitch deck.
Project 1: Use case demonstrator for sustainable glass packaging using upgraded AI capabilities
The use case requirements and objectives in collaboration with an industrial partner, a jam producer have been defined. The jam production site has been visited and the full life cycle of the glass packaging has been mapped out with the partner. This includes all operation, transportation and usage condition. Based on the mapped life cycle three load cases of the packaging have been identified.
A parametric design of the glass packaging has been set up. The glas manufacture we partnered up with provided mechanical test laboratory results, which have been used to validate the digital twin. A design of experiment has been designed to use the digital twin for data generation. The funding has been used for the salaries of the people involved as well as for the purchase of the required simulation software.
In this activity, a large data set based on the digital twins have been generated. This data has been used to develop the machine learning models. This involved using specialized software tools and techniques to analyze the data and train the ML models.
The machine learning based model of the glass packaging has been used to run optimisations, design variation and sensitivity analysis to create multiple 3D design packages with lower environmental foot print, minimal material and minimal energy required for manufacturing. The full end-to-end workflow has been applied and showed that for a specific jam jar an improved glass jar design can save up to 22% material and 11% of annual CO2 emission.
A life cycle assessment (LCA) of the glass jar has been carried out. This involved analyzing the environmental impact of raw material, production, transportation, usage and recycling. The funding has supported the data collection and analysis required for the LCA
Project 2: Upgrading the business plan
Based on market data of the raw material market, market data about packaging manufacturers as well as packaging designers and engineers, a product focus of Digimind has been identified as well as a realistic total addressable market. This has been accompanied by an extended survey of customer/packaging designer, engineers and manufacturer their need in digital software tools and services in optimizing packaging. This has helped to fine tune the product offer of Digimind and define the target customer group for Digimind. It showed that the market interest is especially large in the sector of plastic packaging and recycled materials as well as that an end to end offer including a cooperation with a packaging converter is favorable.
In this work package the business plan has been reviewed and improved based on the activities in work packages 2.1 and 2.2. A good understanding of the market has been used to upgrade the business plan. Then, based on the data a robust financial projection has been developed and subsequently our funding needs have been adapted.
In this work package, the findings, data and information has been synthesised into a fully revised pitch deck for investors. A very high-quality investor pitch deck has been developed, that shows a clear understanding of the market, the target customer and a big vision for the future of the company.
The use case requirements and objectives in collaboration with an industrial partner, a jam producer have been defined. The jam production site has been visited and the full life cycle of the glass packaging has been mapped out with the partner. This includes all operation, transportation and usage condition. Based on the mapped life cycle three load cases of the packaging have been identified.
A parametric design of the glass packaging has been set up. The glas manufacture we partnered up with provided mechanical test laboratory results, which have been used to validate the digital twin. A design of experiment has been designed to use the digital twin for data generation. The funding has been used for the salaries of the people involved as well as for the purchase of the required simulation software.
In this activity, a large data set based on the digital twins have been generated. This data has been used to develop the machine learning models. This involved using specialized software tools and techniques to analyze the data and train the ML models.
The machine learning based model of the glass packaging has been used to run optimisations, design variation and sensitivity analysis to create multiple 3D design packages with lower environmental foot print, minimal material and minimal energy required for manufacturing. The full end-to-end workflow has been applied and showed that for a specific jam jar an improved glass jar design can save up to 22% material and 11% of annual CO2 emission.
A life cycle assessment (LCA) of the glass jar has been carried out. This involved analyzing the environmental impact of raw material, production, transportation, usage and recycling. The funding has supported the data collection and analysis required for the LCA
Project 2: Upgrading the business plan
Based on market data of the raw material market, market data about packaging manufacturers as well as packaging designers and engineers, a product focus of Digimind has been identified as well as a realistic total addressable market. This has been accompanied by an extended survey of customer/packaging designer, engineers and manufacturer their need in digital software tools and services in optimizing packaging. This has helped to fine tune the product offer of Digimind and define the target customer group for Digimind. It showed that the market interest is especially large in the sector of plastic packaging and recycled materials as well as that an end to end offer including a cooperation with a packaging converter is favorable.
In this work package the business plan has been reviewed and improved based on the activities in work packages 2.1 and 2.2. A good understanding of the market has been used to upgrade the business plan. Then, based on the data a robust financial projection has been developed and subsequently our funding needs have been adapted.
In this work package, the findings, data and information has been synthesised into a fully revised pitch deck for investors. A very high-quality investor pitch deck has been developed, that shows a clear understanding of the market, the target customer and a big vision for the future of the company.
The achieved machine learning model for the virtual representation of the glass packaging is beyond the state of arts. This can be used to lightweight all kind of glass packaging that can save material used in the packaging and subsequently cost, energy and CO2.