Skip to main content

IEOR Seminar by Dr. Kabir Rustogi

Title: Using OR and ML techniques together to optimise operations in
the Logistics Industry

Speaker: Dr. Kabir Rustogi

Date, Time: 10th October, 2019, 10.30 AM - 11.30 AM

Venue: Seminar Hall, 2nd Floor, IEOR Building

Abstract: Simply put, operations in the logistics industry involve movement of goods between two locations. At scale, this provides an opportunity to operate more efficiently by aggregating shipments that share some common path. This is achieved by designing a distribution network, where nodes allow for aggregation and segregation of shipments, and edges allow for transportation of goods from one node to another. The problem of designing such a network and moving millions of items through it on a daily basis, while ensuring industry standards of speed are met at the lowest cost, lends itself to several optimisation problems. While a lot of these problems, e.g., location routing problem, vehicle routing/scheduling problem, etc are classical in nature, industrial application of these approaches has remained a problem area due to issues such as inadequate information about cost functions, incomplete ground constraints, variability in load profiles, rapidly evolving physical landscape, etc. Companies often spend a lot of time and resources to manually collect this information, so that optimisation algorithms can be fed the right data. Recent advancements in the field of IoT and ML have enabled us to capture vast amounts of data from the ground and derive intelligence from it automatically. These systems allow us to create a better understanding of the ecosystem in which the optimisation models operate in, thereby allowing organisations to deploy solutions sooner, while ensuring that the output is meaningful and readily acceptable by ground operations. This talk will delve deeper into this area, with case studies of similar systems developed at Delhivery, one of India's largest logistics companies.

About the speaker: Kabir heads the Data Sciences vertical at Delhivery. His work involves building machine learning and optimisation solutions for applications in the field of maps, location intelligence, network design and route optimisation. Kabir comes with years of academic experience in the field of optimisation algorithms. He completed his PhD in the area of Scheduling Optimisation in 2013, following which he worked as a lecturer of Operational Research for several years in the UK. His research was awarded as the best PhD by the British OR Society and was later published as a book by Springer Publications.

More about the Scholar
 

News Category
Date Posted