In this crisis, different new hazards like border closing, warehouse restriction, lack of transport capacity… can appear everyday and have a deadly impact on critical supply chains (medical items, food...) for people. We believe that we can use AI to integrate these hazards in an optimisation tool to recalculate realistic transport plans on a daily basis and to forecast storage allocutions on a weekly basis. We've tuned our already existing algorithms and retrained models during the weekend to achieve that. Let's save lives !

The problem our project solves

  • Covid-19 crisis puts supply chain upside-down by introducing several unexpected types of hazards : border closing, warehouse restriction, lack of transport capacity, suppliers failure ...
  • Supply chains are the backbone of physical society. People won’t heal, industries won’t recover, economies won’t raise up if we don’t fix supply chains !
  • Supply chains and those unexpected hazards make a far to complex system to be fixed only by humans
  • Demand forecast is just the first step. It doesn’t tell you how to operate your supply chain to respond to that demand when everything is broken.

The solution we bring to the table :

We already provide our customers our smart SaaS solution that

  • Help them to collaborate remotely, anticipate and optimize their supply chain, daily
  • With a beautiful and dynamic web interface to collaborate and to share datas, status and decisions with all the stakeholders
  • Thanks to our super fast converging algorithms (within minutes compares to hours) enabling real time operational use (works for strategics decisions too)
  • Our technology is multiple award winning and comes from a 5 years R&D team effort. We use an hybridation of deep learning, graph theory, time series modelling and operational research.

We want to use our unique solution to show that it’s possible to find the optimal way to drive the supply chains under Covid-19 constraints, by anticipating the impact of these hazards. On a daily or weekly basis

During the weekend

  • We ‘ve built a prototype dedicated to the supply chain under covid failure on top of our AI algorithms : Digital twin setup, constrains integration (covid hazards), dedicated local front end, synthetic dataset to simulate FFP mask transportation

  • Constraints : border closing, warehouse capacity & workforce, transport capacity & delay, moving demand

  • Results of optimization : On-time delivery rate, warehouse use rate, total cost, dependency tree & single points of failure, global load of the network (snowball effect), order history for all trucks, trains and pallets

  • Achievement (check the video): We succeeded in simulating the impact of a closing a 3 borders on the transportation of FFP mask. Our prototype can find in seconds the best way to reroute the masks and other medical stuff at a pallets range, the cost of the rerouting and the on-time delivery rate ! We've also simulated the impact of relocating industries in Europe on the supply chain by identifying the "points of failure" and backup them. Finally, we've validated that our prototype is able to handle the huge issue that the end of the lockdown represent for supply chain operators.

The solution’s impact to the crisis

Supply Chains are the backbone of physical society. People won’t heal, industries won’t recover, economies won’t raise up if we don’t fix supply chains Regarding the deadly Covid-19 crisis, it's no more time to play solo. We will overcome it only if we work all together, and if we are able to take fast and common decisions.

Our solution can help to do that. Our SaaS stack is scalable by design to handle that complexity, we can calculate optimum scenarios super fast to facilitate decisions and our interface is collaborative by design to tackle silos we can't afford anymore. We've talked with mentors this weekend, read the featured papers, chatted with other participants : our solution is unique in it's capacity to dig at this level of detail when it comes to fast optimisation and collaborative work. Our impact can be in 3 phases : now by helping to deliver medical stuff, at the end of the lockdown to handle the major snowball effect at the European scale cause by the restart of the economy and finally to simulate the impact of relocation strategy.

Before the pandemia, some of our customers saved tens of millions per years in "normal" conditions. We can guess, regarding today's situation, that we are talking about billions at the European scale by now. But in a first place, we are sure we can help save lives by optimizing medical distributions

Business Model and what's next

Our company already has paying customers. The pricing of our licence depends on the size and parameters of the supply chain. The huge changes already made by the pandemia will increase the need for solutions like ours. The world and the supply chains post-Covid will put this trend higher. Furthermore, another advantage of our solution is to increase the fill rate of the trucks. And by increasing this rate, you reduce the number of trucks needed and the road pollution. We already made our customers save huge amount of CO2. To fully embrace this change of paradigm, we will certainly need to invest in our solution to industrialize what we did during this weekend and go further. Even if that first version is strong enough to bring a lot of value in the next months by itself I'm quite sure that the EU could help us in that way with funding like H2020 projects.

Built With

+ 7 more
Share this project: