MODELLING FARM MACHINERY SELECTION FOR SCATTERED FARMS USING MINIMUM-COST METHOD

SOURCE:

Faculty: Engineering
Department: Agricultural And Bio-resources Engineering

CONTRIBUTORS:

Amaefule, D. O.
Oluka, I.
Nwuba, U.

ABSTRACT:

Farm mechanization increases agricultural production but is rendered uneconomical and impeded when the farms are small-sized and scattered. Such farms have received little machinery selection attention. This study models machinery selection for fragmented scattered farms as a pool farm so as to afford their mechanization. Field operation labour cost was added to the Hunt-Wilson’s annual machinery cost equation. Inter-farm machinery transport costs were incorporated into the equations, thereby enabling its application to both small and large noncontiguous pool farms. New minimum-cost machinery size selection models were developed based on Hunt and Wilsons (2015) principles. With the new models it is now possible to circumvent the use of a prior arbitrary width-dependent machinery capacity required in the Hunt-Wilson (2015) least-cost width selection model. A model for adjusting the selected field capacity so as to account for the transportation time loss was also developed. The developed models were validated with parameters from tillage operations on the farm lands serviced by the Anambra State Ministry of Agriculture tractor hiring outfit Awka, Nigeria. The farms totaled 675 ha. The annual machinery costs per hectare predicted by the models had very
close values at each farm size when the considered farm size exceeded 420 ha. Comparatively, the developed model with no operation labour cost-consideration (denoted as Z- model) selected the lowest minimum-cost implement size. The Hunt-Wilson model (denoted as H- model) selected the largest for most of the pool farm sizes considered. The tillage machinery and transport cost per hectare incurred was generally highest for the H- model and least for the developed model with operation labour cost-consideration (denoted as L- model). The models were sensitive to farm size and required Farm size and fuel, tractor and implement prices, and tractor power as inputs. The minimum-cost plough width predicted by the L- model for example increased from 0.5481 m to 0.9740 m, 2.1009 and 2.5995 m for pool farm sizes varying from 45 ha to 145 ha, 420 ha and 675 ha. The Z- model plough width predicted was 0.2251 m, 0.4242 m, 1.5880 and 2.0579 m and the H- model width 0.8186 m, 1.0201 m, 1.5880 m and 1.7587 m, when compared with the L- models widths in the same order. The ANOVA of the widths from the 3 models showed no significant difference at a 0.05 level of significance. The incurred cost per hectare with the L- model decreased from N12,485.44 to N8,480.44, N7,611.25 and N6,892.56 for the same farm sizes in the same order. The cost was N16,470.60, N10,452.75, N7,650.97 and N6,891.91 for the Z- model, while for the H- model it and was N12,378.38, N8,471.28, N7,708.13 and N7,010.44 for the listed farm sizes in that order. The foregoing showed that pooling small farms from 45 ha to 675 ha for combined equipment use could reduce the considered machinery cost per hectare to nearly half of the value. This agreed with the well-known fact that it is more economical to mechanize bigger farms. Finally models that take advantage of the non-coinciding field operation time windows of individual farms to reduce the nominal pool sizes to smaller ones were developed. They reduced the simulated 675 ha nominal farm size to an equivalent 430 ha size for the tillage implements capacity selection. This will reduce the required tillage machinery capacity and corresponding total annual machinery cost. The pieces of machinery recommended following the prediction on the basis of for this equivalent 430 ha pool farm were two 0.3100 ha/hr (0.90 m width) disc plough, one 0.7507 ha/hr (1.70 m width) disc harrow and one 0.5702 ha/hr (1.80 m width) disc ridger. They are to be powered by two 48.5 kW, one 61.1 kW and another 61.1 kW tractors
respectively, based on the available tractor and implements in the local market. It is hoped that the application of the developed models will enhance the mechanization of small scattered farms in a cost-effective way.