Abstract:
The crankshaft is the part of an engine which translates reciprocating linear piston motion into rotation. Automobile crankshaft failure is caused by age and distance covered, among other factors. The need to determine the age and distance to be covered before failure is very important, hence the development of this model for proffering solution to this problem. These attributes which caused crankshaft failure were determine using questionnaire administration and oral interviews based on expert opinion and experience. The three selected crankshaft maintenance shops in Akure
and Ondo, Nigeria are: Afolabi Engineering Company Akure (Workshop A), Danzaki Technical Company Akure (Workshop B), and Twins Engineering Company Ondo (Workshop C). The selected automobile vehicle brand/models are: Toyota Camry, Mazda 626, Mercedes 230 C Class, Honda Civic, Peugeot 607 and Ford C100. Failure rate and failure consequence threshold marking were determined in order to arrive at manageable vehicular operation system based on decision rules namely: reconditioning process, replacement and interchangeability system of the crankshaft components where the optimum failure rate (R1) and failure consequence (β1) occurred. Failure
rate based on age and distance travel variation of the vehicular crankshafts was modelled using multivariate linear regression approach. Computer algorithm was developed for the software model using Microsoft Visual C# computer language. The model and its software were tested to determine their level of performance. The results generated from the application of these models were categorised into six scenarios whereby the Mercedes 230 C Class gave the best result, its failure rate was difficult to identify up till the tenth 10 years of 1st scenario. Considering the failure consequence under this scenario 1, Toyota Camry generated the best allowable income per day for
the workshop which is N49,000 per day. The paired T-test was used to check the level of 5% level of significant between the mean of the actual values and the mean of the predicted values for failure rate and failure consequence which showed that there are no significant difference between the variables at 5% level of significant. The model and the software will be veritable tools in solving crankshaft failure problems in automobile, manufacturing, and machine tools industries.
