Toward Reducing Crop Spoilage and Increasing Small Farmer Profits in India

Abstract

India's agricultural system has been facing a severe problem of crop wastage. A key contributing factor to this problem is that many small farmers lack access to reliable cold storage that extends crop shelf-life. To avoid having leftover crops that spoil, these farmers often sell their crops at unfavorable low prices. Inevitably, not all crops are sold before spoilage. Even if the farmers have access to cold storage, the farmers may not know how long to hold different crops in cold storage for, which hinges on strategizing over when and where to sell their harvest. In this note, we present progress toward a simultaneous hardware and software solution that aims to help farmers reduce crop spoilage and increase their profits. The hardware is a cost-effective solar-powered refrigerator and control unit. The software refers to a produce price forecasting system, for which we have tested a number of machine learning methods. Note that unlike standard price forecasting tasks such as for stock market data, the produce price data from predominantly rural Indian markets have a large amount of missing values. In developing our two-pronged solution, we are actively working with farmers at two pilot sites in Karnataka and Odisha.

An interpretable produce price forecasting system for small and marginal farmers in India using collaborative filtering and adaptive nearest neighbors

“…(1) the raw accuracy, which is calculated as the fraction of y m,s that is correctly classified, and (2) the "balanced" accuracy (Chen et al, 2017), which is calculated as the average of the raw accuracies in classifying each of the three price change directions y m,s ∈ {1, 0, −1} (i.e., we calculate the raw accuracy confined to each of the three price change directions and then we average these three fractions).…”

“…None of the above studies predict produce prices for multiple time periods and multiple markets, and their prediction results are not presented in a fashion that is easy to interpret and explore by farmers. The closest work to this paper is that of Chen et al (2017), which benchmarks a few machine learning algorithms forecasting price change for the next day looking at 14 markets and one specific produce.…”

“…As produce price forecasting somewhat resembles stock price forecasting, some studies view produce as special stocks with high volatility, low frequency, and low trading volume (Brandt and Bessler, 1983;Kumar, 2004;Chen et al, 2017). However, produce markets differ from stock markets in a few key aspects:…”

“…Related work To the best of our knowledge, only a few studies work on produce markets. Brandt and Bessler (1983) explored time series models and expert models to predict produce prices for the next quarter, and Chen et al (2017) predicted the price trend in the near future with a number of machine learning methods. Chakraborty et al (2016) predict food prices using both Agmarknet data and news events with the idea that some news events (e.g., worker strikes, festivals) can help predict price shocks.…”