PICASA: Predictive Interventions in Capacity Allocations through Systemic Automations
Identifying and providing differentiated experiences to customer segments is crucial to building sustained customer engagement. At Flipkart, one of the approaches used to provide differentiated experience is by creating dedicated processes that are customized for the said experience. At the same time, capacities allocated to these dedicated processes should not be underutilized. This requires a balance between providing differentiated experiences and capacity utilizations. In this talk, we will present PICASA, an automated system that makes predictive interventions to optimize capacity allocation across dedicated processes. This system enables us to provide an improved experience to more customers from the available capacity while also providing a differentiated experience to key customer segments.
- Introduction to the business problem and technical need
- Brief highlight of the existing process and it’s challenges
- Our approach to solving the problem
- Results and Insights
Gowtham Bellala is a senior data scientist at Flipkart. He is focussing on Supply Chain Automation and Optimization by leveraging Machine learning, Artificial Intelligence and Big data. He holds a PhD in computer science from the University of Michigan, and is passionate about data science, machine learning and privacy. In the past, he has worked at C3.ai, HP Labs and GE Research on the application of these techniques in Healthcare, Sustainability and the Industrial Internet of things.