International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 05 | May 2022
p-ISSN: 2395-0072
www.irjet.net
Analyzing Cost of Crop Production and Forecasting the Price of a Crop Swati Shirsath1, Prof. Sanjay Pingat2 1Department
of Computer Engineering, Smt. Kashibai Navale, College of Engineering, Pune, India of Computer Engineering, MKSSS’s Cummins College of Engineering for Women, Pune, India 2Department of Computer Engineering, Smt. Kashibai Navale, College of Engineering, Pune, India ---------------------------------------------------------------------***--------------------------------------------------------------------2. Related Work Abstract - Various technologies such as the Internet of 1Department
Things, Artificial Intelligence, Drones, and others have been used to automate the agricultural sector in recent years. Because the weather is constantly changing, managing finances is a major concern in a country like India, and water is not always freely available, automation and precision are essential in this era. Researchers have come up with a slew of solutions, including weather forecasting, smart water management, inspecting crop quality, identifying disease, spraying pesticides on diseased crops, and so on. Machine learning, artificial intelligence, the internet of things, sensors, and other technologies are commonly used in these solutions. All of these solutions are intelligently managed and provide a high level of output and precision to assist a farmer in increasing productivity. Finance is the most important issue among these, yet it is not addressed by scholars. Only a few academics have used time series analysis to anticipate agricultural prices. And this predicting is limited to specific crops, vegetables, and fruits, as well as a specific region (s). There is no global mechanism for forecasting crop prices. We conducted a comprehensive evaluation of numerous strategies for predicting/forecasting agricultural prices, including various features, procedures/methodologies used, and their accuracy. We also provide research on the cost of crop production, why the cost of crop production is necessary, software available in the market for the purpose, formulae used in crop production calculations, elements considered in crop production calculations, and so on.
2.1 Predicting crop price A lot of research has gone into forecasting the yield of a crop based on numerous factors such as weather, soil type, soil moisture, and so on. After that, a few scholars worked on predicting crop prices. The practice of forecasting crop prices (primarily) using time-series forecasting is known as crop price prediction. The majority of the academics focused on predicting/forecasting the price of a given crop. And the majority of them display results for their own country or region. Crop price forecasting at the global level has yet to be completed. Prediction/forecasting of specific crop(s) price is critical not only for farmers (to yield specific crops for a specific region), but also for regional governments and other stakeholders such as supply chain marketers. Prediction/forecasting of specific crop(s) price is critical not only for farmers (to yield specific crops for a specific region), but also for regional governments and other stakeholders such as supply chain marketers. Tomato and maize (only for the state of Karnataka, India) [18], rabi and kharif [20], Arecanuts [21] (just for Kerala city of India), cabbage, bok choy, watermelon, and cauliflower [19] (for the capital of Taiwan, Taipei), tomato, onion, and potato (TOP crops, for India) [2]. Various models, such as univariate and multivariate models, have been proposed [1]. Exogenous variables such as VAR model [21], structured VAR model [22], ARIMAX, SARIMAX, and endogenous variables such as MA, ARMA, ARIMA [3] AND exogenous variables such as VAR model [21], structured VAR model [22], ARIMAX, SARIMAX. Researchers have proposed machine learning models such as support vector regression [4], LASSO [5], LSTM [6,7], ensemble models [20,24,25], and others.
Key Words: AI, IoT, CNN, Crop production Cost, Crop Price Forecasting, Image processing.
1. INTRODUCTION In reality, crop production costs are determined by a variety of elements, and if a digital platform offers such applications, accuracy is critical. After testing multiple strategies on a similar dataset, the accuracy of crop production cost prediction can be improved. As a result, the technique chosen is critical, as farmers' cost calculations cannot be as precise as those generated by an IoT-based system. Until far, scientists have focused their efforts on estimating the price of a certain crop. It has not yet been calculated using image processing and CNN. We propose a system that uses image processing or sensors to recognize various soil properties such as fertility, moisture, pH value, and so on, and then processes the data in the cloud to compute crop production costs.
© 2022, IRJET
|
Impact Factor value: 7.529
[18] focuses on projecting tomato and maize prices in the Indian state of Karnataka. Trend analysis was employed as the methodology. Historical pricing, market arrival quantity of crops, historical weather data, and data quality-related factors were all utilized in this study. They concentrated on projecting agricultural prices while dealing with high price swings. They are more accurate than traditional forecasting techniques. [19] is concerned with forecasting cabbage, bok choy, watermelon, and cauliflower prices in Taiwan. The most important element in this work was the utilization of historical data to automate the forecasting procedure for the
|
ISO 9001:2008 Certified Journal
|
Page 198