Name: Aniket Deo
Name of supervisor: Prof. Amit Arora and Prof. Shubankar Karmakar
Description of research work:
Indian agriculture has an unfortunate history of farmers suicides, indebtedness and heavy migration of agricultural manpower into other sectors. The reason is justified by successive crop failures, increasing inputs costs, market rate uncertainties and vicious climate change. On the contrary to the declining workforce in agriculture, the demand for food in India is constantly rising. Moreover with income growth, urbanisation and trade liberalization, the dietary habits of the Indian population has diversified putting an additional pressure of crop diversification on the farming systems and their supply chains. Most of the farming systems in India operate under resource scarcity given their small land holding and poor accessibility to irrigation and credits. In such a scenario, production has to be achieved with efficient resource management such that the economics of farming is profitable and sustainability of resources for next seasons is ensured.
Our work focusses on designing appropriate production plans for optimal resource management such that maximum gains are generated from the available resources. An extensive field level investigation was carried out in a tribal cluster of Palghar district, Maharashtra to study local farming systems and existing production plans. An observation was made that farmers judgement about production planning has been failing recently given uncertainties in the external environment of the farming systems. A study to access their existing production plans suggested that farmers are misjudging their resource capacities while planning their production leading to crop/yield failures, low returns and poor resource use efficiency. It is understood that decisions about production planning involves complex interactions between biological, economic and social factors in farming system. To address this complexity in decision making, we have developed a micro-planning tool that suggest farmers with options of appropriate production/cropping plans considering their agro-socio-economic profiles. The current study poses novelty in integration of analytical and investigation tools to address a real-life problem. The model incorporates social, ecological and economic aspects considered by a farm-household during decision making. A multi-stage procedure is developed to generate full factorial crop combinations using combination algorithm, allocate optimal acreage to crops using linear programming to form cropping plans, clustering/ grouping of all cropping plan alternatives using the K-means algorithm, ranking clusters based on multi-criteria decision-making tool TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) and uncertainty analysis of Rank-1 (top) cluster via Monte-Carlo simulation to suggest a list of appropriate cropping plan options. These options provide flexibility of choices which strengthen the model in comparison to the existing models while the plans suggested by the model are most profitable and least risky.
Several case studies are analysed through this model. For a case study wherein the farmer practiced 4 crop-mix system and had dilemma in choosing among 15 crops, the model generated 1365 options of production plans among which 12 plans were filtered as most appropriate. Figures 1 and 2 describe the model designed production plans. The plan recommends land use distribution among the crops (figure 1) and choice of crop at particular order (figure 2).
The model developed in this study is a valuable tool to explore all feasible options for production. The farmers could use this model to minimize loses and optimize resource allocations. A preliminary analysis in the ongoing study suggests that this model can increase the farmer’s income by around 22% while mitigate the risk by almost 28% by just changing their crop selections and acreages. The application of this model extends to policymakers/planners to evaluate the utility/efficacy of the existing cropping systems of farmers. The model can improve the mental models/heuristics of farmers regarding their farming systems, empowering them to make appropriate decisions. The comprehensive model framework is generic and valid at multiple scales, and scenarios. This exploratory study provides unique and convenient framework for designing more robust production planning models, which would include complex social (farmer's behaviour and heuristics), biophysical (biotic and abiotic components), and economic relationships and uncertainties in the farming system.
Figure 1: Land use distribution of most appropriate cropping plans
Figure 2: Percentage of occurrence of crops respective to orders