Analysis of Agricultural Policy Recommendations using Multi-Agent Systems

TL;DR

This paper models a segment of India's agricultural system as a Multi-Agent System to analyze policy impacts, aiming to inform better policy decisions and reduce farmer distress.

Contribution

It introduces a novel multi-agent model of Indian agriculture that incorporates water, market, and information factors for policy analysis.

Findings

Identified potential negative impacts of certain policies on farmers.

Simulated policy scenarios to predict outcomes in specific regions.

Recommended policies based on simulation results.

Abstract

Despite agriculture being the primary source of livelihood for more than half of India's population, several socio-economic policies are implemented in the Indian agricultural sector without paying enough attention to the possible outcomes of the policies. The negative impact of some policies can be seen in the huge distress suffered by farmers as documented by several studies and reported in the media on a regular basis. In this paper, we model a specific troubled agricultural sub-system in India as a Multi-Agent System and use it to analyse the impact of some policies. Ideally, we should be able to model the entire system, including all the external dependencies from other systems - for example availability of labour or water may depend on other sources of employment, water rights and so on - but for our purpose, we start with a fairly basic model not taking into account such external effects. As per our knowledge there are no available models which considers factors like water levels, availability of information and market simulation in the Indian context. So, we plugged in various entities into the model to make it sufficiently close to observed realities, at least in some selected regions of India. We evaluate some policy options to get an understanding of changes that may happen once such policies are implemented. Then we recommended some policies based on the result of the simulation.