Agricultural mechanization is essential to agriculture, but during its practice, tractor operators are still exposed to probable stress agents, which depress their work efficiency. The present study was conducted to formulate a representative multivariate mathematical model of the occupational comfort index of an agricultural tractor, based on thermoacoustic indicators, in the periodic soil preparation. The study was carried out on 4.3 ha, using a mechanized set without an ergonomic cabin, coupled with the respective implements, for plowing, harrowing and furrowing operations, for a daily workday of eight hours, performed by the same operator. Data records on the work platform were as follows: air temperature (°C), black globe temperature (°C), relative air humidity (%), operator's heart rate (bpm), head peripheral temperature (°C), arm peripheral temperature (°C), noise (dB) and wind speed (m s−1). Data were subjected to exploratory analysis using the principal component analysis (PCA). The occupational comfort index (OCI) prediction model was created using multivariate linear regression analysis. The OCI validation achieved excellent performance, based on the high values of agreement (> 0.90) and confidence (0.86), which allowed its use to characterize the physical workload and unsanitary conditions for agricultural tractor operators.