Our case studies
Out of our own experience in the food production sector we saw that especially SMEs have a hard time to implement a Quality Management System with their limited ressources at hand. We started a focus group in April 2015 with farmers, producers and interest representatives of the food sector in Austria. Researching and talking with pears helped us to really get to the point of all evil here: complexity (Kleboth, Luning and Fogliano 2016).
We visited more than 20 farmers and 30 producers in Austria, Germany and Switzerland. We followed their work and talked with them to get a feeling of what they would need to deal with the complexity in the implemention of a Quality Management System and its documentation. Next to having a blast with the food producers we got invaluable insights into the real needs of the food producers.
The interpretation of the results of our empirical on-site and off-site research was done in over a periode of more than six month. We discussed our findings with the food producers, researchers and other stakeholders. Finally we defined what we call the „ressource and complexity challenge“ as the problem we have to solve.
In this phase we explored quite some different technologies and approaches: From system dynamics modelling to machine learning and artificial intelligence. The last one is the one we sticked with.
We sat down and started coding. First a low proof-of-concept prototype to simply proof that our intelligence was working. From there we developed a minimum-viable-product to be able to test it in real life.
This is the phase we are in right now. We have beta-tester who work close together with us to test the new platform and give us feedback. Such a great feeling to see it in action and making real impact!
The “sumo.platform” solves more foundational – long term problems of the food sector. In our own life as food quality managers and developers we run constantly into the same problem: lack of data. lack of prediction power. lack of insight power. The sumo platform is our solution to this problem. Bam!
One problem we always run into was the lack of good and sufficient data for the food sector. When you want to use the latest technology then you need data. A lot of it! Interesting enough: we gather massive amounts of data. What is the problem here? Mistrust, Security and lack of data comparison tools seemed to be obvious.
We invited other food related data scientists, data analysts from food companies and software developers for a workshop. We discussed the obstacles and problems when dealing with data and the lack of it. We saw that everyone had the same problem.
In this phase we wrote a internal white paper called „42 – the answer to everything“. We stated that a single-data-one-stop-shop is not available but desperately needed. We formulated the problems and why we think it is important to solve them: Data-matching, pre-trained data and compliance.
Together with peers we sat down and discussed possible solutions. With the whitepaper as basis we worked us through every identified problem-point. We evaluated different technologies, business-models and acceptance hypothesis. Then we started…
… building our proof of concept. In this phase we could show that a sophisticated set of algorithms is able to make sense of data with even a little fingerprint of it. Exciting times!