Transformer models specially ChatGPT have democratised AI. With the advent of ChatGPT, most of the small to mid size firms have started looking out for answers to their business in AI. It has just started to make people realise how AI can solve for most of the business inefficiencies across industries.
Traditionally Supply Chain has been one of the most complicated area of business. Humans have burnt out themselves solving supply chain problems since a long past and the complication never ends. Be it CPG, FMCG or Retailers, everyone needs to do a very robust demand, inventory, pricing and promotion planning. It is still possible to do the planning well for your top SKUs during holidays or promotional time, but planning for entire portfolio on a business as usual day is a great challenge, depending upon scale.
This hack session will involve talk around designing scalable systems for solving different supply chain problems in an inter-connected fashion using AutoML solutions. It first goes into the problem details and thereby getting into the solution design of different problems in an automated fashion using Machine Learning, Deep Learning architectures and optimisation.
The session elaborates different neural architecture designs for solving supply chain optimisation problems. It also explains how and what part of business are learnt by which neural layers. As an example, how we do we make our Neural model learn more complicated promotions.
Key Takeaways
- Burning Supply chain problems across industry
- Designing AI systems in an AutoML fashion for solving certain supply chain problems in an inter-connected manner
- Deep dive over different Neural Architectures, Optimisations and inequalities for solving supply chain problems
- Learning to design neural layers which can learn specific details of business in more/less complicated fashion.
