Programme

PGP

Term

IV

Academic Year

2021-22

Course title

Working with Networks

Area

Production & Quantitative Methods

Credits

1.00

Instructor(s)

Prof. Diptesh Ghosh

Course Description & Objectives

Introduction:   This is a course on analysis of network structures. Whenever entities are interconnected, their interconnection can be represented in structures called networks. These structures are abstractions consisting of nodes which represent the entities, and edges that represent the connections.  Networks are ubiquitous in present day life. For example, supply chains can be visualized as networks. The diagram below represents a typical supply chain network for a pharmaceutical company.   Trade relations can be also be represented as networks. The next diagram is a network representation of the way in which sugar was traded around the world in 2010-11. (Source: cdn3.vox-cdn.com) Given these networks, and properties of the entities of the network, one may ask interesting questions. For example, what is the maximum amount of products that can be sent from one point to another through these networks? Or how sensitive is the chain to the disruption of one of its entities (i.e., nodes or connections)? Interstate migration can be represented by networks. The diagram below shows a network representing migration patterns among different states in the US in 2007. (Source: A. S. Chakrabarti and A. Sengupta, Econ. Mod. 2017)  Based on the data, one can ask questions about the scale of migration in an economy and predict about the populations that individual states may assume after a given time if these patterns continue. There could be questions on the costs and risks of such migration, and the effect of particular government policies on these patterns.  Even one’s social media presence can be visualized as a network. The next diagram is a visualization of how my Facebook friends are interconnected on Facebook using the Lost Circles app for Chrome. Interesting questions here would be to identify the central individuals are in the network, and how close individuals are to other individuals in the network, and to identify communities in the network. Objectives:  The present course equips participants with tools to analyse the structure of networks and algorithms for effective decision making about problems posed on networks.

Pedagogy

The course will be lecture based. We will also be modeling the problems being discussed and developing computer code to implement the algorithms discussed.