Mathematical models to predict repair and maintenance cost for 2WD tractors in the mechanized rainfed areas, eastern Sudan

Lotfie A. Yousif

Abstract


Many tractor makes are working in the mechanized rainfed areas of eastern Sudan. Most of them are two-wheel drive (2WD) tractors. However, information concerning their repair and maintenance (R and M) costs are inadequate. Moreover, depending on information on R and M costs from other farming systems or countries is unrealistic. Four tractors were studied, which were Massy Ferguson (MF-290), Ford (FD-6610), Belarus (BL-800) and New Holland (TT-75). The objectives were to develop mathematical models to estimate tractors' annual R and M costs, to rank the studied tractors according to their predicted R and M costs and to compare estimations of the general model developed from this study with models developed elsewhere. Data on purchase price, annual working hours, fuel consumption, R and M costs, labor wage and oil costs were collected. The regression and F test analyses methods were used. Results revealed that the power function Y= aXb can be used to estimate tractor R and M costs; where Y is R and M costs as percent of purchase price and X is annual working hours. The annual working hours had very highly significant (P= 0.001) effect on R and M costs. The values of coefficient of determination (R2) were above 0.95 for all tractors and their general model. The developed models for all tractors were arranged in ascending order according to their predicted R and M costs as follows; Y = 0.05 (X/100)1.3 for TT-75,                    Y = 0.04(X/100)1.4 for BL-800, Y= 0.03 (X/100)1.7 for MF-290 and Y = 0.05 (X/100)1.5 for FD-6610. The general model from this study was                       Y = (0.041 (X/100)1.43). This general model predicted lower R and M costs compared to the models suggested for irrigated schemes of Sudan and some other countries. It is advised that each specific area and country must develop its own mathematical model.


Full Text: PDF

Refbacks

  • There are currently no refbacks.