Look Who it Was Random video

Sales fleet transformation and optimisation

303 views · Published 26 November 2018 · 34:10 · Indexed 12 October 2026

Channel: Hasgeek TV · 2018 · Science & Technology

Watch on YouTube

Sales fleet permanent journey planning is a standard practice for industries that utilise a push sales model. The decision making process involves assignment of a set of outlets/stores to salesman to visit on a particular day. The job of the salesman is to cover all outlets in the designated time, and maximise sales. Although, sales fleet management has seen a number of automation solutions, almost no solutions are available for optimal assignment of salesman to outlet to day such that total costs are minimised and revenues are maximised.

The talk is on use of mathematical programming, heuristic optimisation and their integration with machine learning models to create optimal journey plans for some leading FMCG companies in India and South-east Asia. It also talks about numerous real world challenges around deploying the plans on ground, and the need for a deep understanding of the ‘human element’, while designing optimisation solutions for the real world. The talk will also briefly explain the future trend of the push sales model in FMCG, and its possible transformation into a multi-channel model.

The talk is structured in four major parts. The first part will talk about - what goes in organising sales fleet movement?. It will also give some qualitative insight into problems of the stakeholders. The second section will talk about quantitative insight of the problem. In the third section, we formally describe the problem, talk about the solution approach and show real world examples of implementations. In the final section, we talk about future trends.

1. What is a sales beat?
2. Why a sales beat?
3. Stakeholder point of views - Salesman, Shopkeeper and Sales Managers
4. Bird eye view into the problem
5. Real world visualisations of the problem
6. Formal problem statement
7. Complexity and scale of problem
8. Solution approach
9. Real world visualisations of the solution
10. Changes and implementation challenges
11. Way forward
12. Summary and conclusion

Rahul is a graduate from UC Berkeley and has experience across diverse fields ranging from Mechanical Engineering, Operations Research to User Centered Design and Product Development. He has expertise in mathematical modelling techniques like mixed integer programming, heuristics, meta-heuristics and stochastic programming. In the past two years, he has focused on developing novel optimisation algorithms for routing, network optimisation and inventory problems at Locus.sh. He also has expertise in building machine learning models and integrating them with classical optimisation techniques.

More from this channel