This content is not included in your SAE MOBILUS subscription, or you are not logged in.

A Telematics Enabled Analytics Approach for Determining Tractor Usage

Journal Article
2021-26-0089
ISSN: 2641-9637, e-ISSN: 2641-9645
Published September 22, 2021 by SAE International in United States
A Telematics Enabled Analytics Approach for Determining Tractor Usage
Citation: Natteri, A., Surendran, J., PAUL RAJ BOB, B., Pushparaj, K. et al., "A Telematics Enabled Analytics Approach for Determining Tractor Usage," SAE Int. J. Adv. & Curr. Prac. in Mobility 4(3):734-741, 2022, https://doi.org/10.4271/2021-26-0089.
Language: English

Abstract:

Tractor lease is an attractive proposition for farmers with small land holdings in India as initial investment required for purchasing a tractor is high [1]. The tractor is wet leased on a daily basis with the driver paid by the hour. Thus, there is a natural tendency by the driver to prolong the operation by taking frequent breaks adding to the overall input cost for the marginal farmer. Therefore, there is need to monitor these operations in real-time to ensure maximum utilization of tractors. The advent of connected and data driven technologies have positively disrupted several sectors including agriculture [2]. Vehicular and GPS (Global Positioning System) data from connected tractors powered by telematic devices can be effectively used for monitoring tractor’s health and position in real time using a mobile application. Moving beyond real-time monitoring, data obtained from connected tractors allow the computation of total field area and on-road distance covered during the day. These metrics enable usage-based pricing to pay for both driver’s labour and tractor rental thereby providing significant cost savings to farmers. The computation of field area and on-road distance covered is a challenging problem as it requires classification of geolocation data into field work and on-road travel points without the use of sophisticated sensors or user inputs. This work presents a telematics enabled analytics approach for monitoring tractor usage. The analytics module comprises of an algorithm that uses a clustering technique taking key vehicle parameters along with GPS data for determining the usage pattern. A polygon-based approach for area calculation is adopted that works for fields with irregular shapes. A novel feature to auto detect & deduce area of large untilled area patches enclosed within the field is also developed. The paper also highlights methods developed to reduce misclassification of GPS points and estimate usage during network connectivity issues.