Advanced search
Start date

Picsel: a new platform to protect rural farmers against weathering weather via crop insurance: from insurance quotes to claims management


The first half of 2021 was characterized by severe weather conditions that affected millions of producers, especially small and medium-sized ones. The occurrence of severe drought followed by two occurrences of frost reduced the value of corn production in the State of Paraná by R$ 11.3 billion, while in the State of Mato Grosso do Sul the reduction was R$ 7.7 billion. Coffee in the state of Minas Gerais presented losses of around R$ 9 billion. In turn, sugarcane suffered the worst crop failure in the last ten years, with an expected reduction of 5 million tons of sugar and 3 billion liters of ethanol. Despite the relative importance of agribusiness in GDP, there are several risks that hinder the stability of the sector's income, especially for small and medium-sized rural producers. According to the report by the World Bank and Embrapa, drought is one of the most important. The problem is that the vast majority of producers do not have access to agricultural insurance. About 4.9 million rural producers, around 98% of the total, are completely unprotected. The main reasons are: i) strong information asymmetry caused by the lack of risk data for rural properties; ii) products that do not adhere to the real needs of producers; iii) high cost of insurance, even after the Government subsidy; and, iv) limited operation and territorial coverage of insurance companies, due to lack of data; and, finally, v) the low incorporation of new technologies and the high bureaucracy throughout the insurance chain. To a greater or lesser degree, all problems are related to lack of data and analog processes and procedures. The main data source currently used by the insurance market is based on municipal productivity estimates from the Brazilian Institute of Geography and Statistics (IBGE). However, as they are measured at the municipal scale, these data do not correctly reveal the risk at the rural property level. In addition, the georeferenced data of the properties are little used, there is ignorance of the planting and harvesting dates practiced in these areas, difficulty in confirming the cultivation in the area, low use of remote monitoring (satellite images and climate information) and poor control of inspections of claims. The solution proposed in this project aims to overcome scientific, technological, operational and commercial challenges, by expanding and improving the results obtained in PIPE Phase I and extending the scope of the project to cotton, coffee and sugarcane crops. The proposed solution is a platform to quote, contract, monitor the insured areas and manage claims through the data generated by the algorithms and the intelligence developed in this project. On the platform, rural producers will find products that are much more compliant with their needs at a relatively lower price. Insurers will be able to improve the risk selection and classification process (underwriting), remotely monitor all insured properties and anticipate possible claims, in addition to managing claims more efficiently and ensuring quality, safety and transparency in inspections, avoiding errors and fraud. At the end of the project, it is intended to expand the access of producers to insurance, mainly for small and medium producers. All rural properties can be located and analyzed by their historical series of productivity generated by our algorithms, their quantified and correctly priced risks, their protection needs precisely adjusted to the products, their properties monitored, and in the event of a claim, their losses quickly assessed and the payment of the indemnity, which currently takes up to 60 days, on average, can be made right after the approval of the inspection report by the insured, that is, in a few days. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items

Please report errors in scientific publications list by writing to: