F. Afsharnia; A. Marzban
Abstract
Introduction Optimal operation and maintenance of engineering systems heavily relies on the accurate prediction of their failures. Repairable engineering systems are well known in industries. A repairable engineering system indicates that the performance of this system after each failure can be restored ...
Read More
Introduction Optimal operation and maintenance of engineering systems heavily relies on the accurate prediction of their failures. Repairable engineering systems are well known in industries. A repairable engineering system indicates that the performance of this system after each failure can be restored through suitable maintenance. It is normally a complex system composed of a number of components. Failure prediction of a repairable system and its subsystems is an important topic in the reliability engineering. One of the most important repairable systems in agro-industrial companies is the sugarcane harvester. This machine has a key role in harvesting operations of sugarcane plant. The failures of this machine causes delay in operations and reduce products yield and quality. Currently, preventive maintenance is conducted on these harvesters to improve the overall reliability of these systems. Therefore, in this study, the long-term effect of preventive maintenance strategy on the efficiency and failure rate of the sugarcane harvester was investigated. Materials and Methods This research was carried out on 30 sugarcane harvesters used by sugarcane and by Products Development Company of Khuzestan during 6 years period. The goal of this study was to introduce a methodology aimed to acquire the information to predict the effect of preventive maintenance strategy on the failure rate and efficiency of sugarcane harvester by time series. Time series forecasting is the use of a model to predict future values based on previously observed values. The expected shape is a forecast from a combination of an ARIMA models (AR, MA, ARMA and ARIMA). The first step in analyzing the time series is plotting the data and obtaining the sample records. The next step is consideration of a trend and periodic components and remove them from the time series and fitting the static model on the time series. The next stage is implementation of the data normalization using skewness coefficient method and their normalization through logarithm differentiation of data. The arithmetic mean of data was applied to obtain zero average of the time series. Sample ACF (Auto Correlation Function) and PACF (Partial Auto Correlation Function) was drawn and then the model rank "a" was recognized and selected by comparison of ACF and PACF for AR, MA, ARMA, and ARIMA models. Results and Discussion According to the results of failure rate related to the sugarcane harvester, it can be found that the mean failure rate of this machine for the 6-years period was equal to 0.147 per hour. The minimum and maximum value of the failure rate were 0 and 0.517 per hour, respectively. The mean annual use hours of these harvesters was 189.8 h. Although the accumulated used hours increased, the mean time between failures (MTBF) was increased. According to Jacobs et al. (1983), the machines may breakdown due to a design defect, physical damage, or normal wear and tear, but many times machines fail because of a neglect and the lack of properly scheduled maintenance. In this study, implemented preventive maintenance resulted in decreasing of failure rate and increasing of machine efficiency as well. In 2016, the failure rate of sugarcane harvester was decreased by 73.23% and the machine efficiency was increased by 14.9% compared to 2011, because timely preventative maintenance and inspection will not only help to reduce major problems and downtime, but it will also help to identify problems when they can be corrected with relatively minor repairs. Among the 12 studied subsystems, topper, electric and motor subsystems were more affected by preventive maintenance by 94.75%, 80.46% and 58.74% decreasing in the failure rate, respectively. With regard to the evaluation criteria such as AIC, MAPE and RMSE, the ARIMA (1, 3, 2) model was determined as a suitable model for predicting the failure rate of sugarcane harvester. Furthermore, there is no significant difference between statistical descriptive measures of forecasting and actual tractor failure rate that it represents high accuracy of forecasting via ARIMA model. Conclusion This study was adapted to the preventive maintenance as a useful strategy that could increase availability and operational efficiency of the sugarcane harvester. Furthermore, it focused on time series modeling method to analyze and forecast the reliability characteristics such as the expected number of failures per interval (failure rate). It is found that time series model is a viable alternative that gives satisfactory results for interval failure predictions in terms of its predictive performance for the sugarcane harvester reliability.
F. Afsharnia; A. Marzban
Abstract
Introduction Given the risk management and improving the process, reliability is important in operations and production management, especially agricultural process. Failure modes and effects analysis (FMEA) is regarded as one of the most powerful methods in this area. High applicability and proper analyzability ...
Read More
Introduction Given the risk management and improving the process, reliability is important in operations and production management, especially agricultural process. Failure modes and effects analysis (FMEA) is regarded as one of the most powerful methods in this area. High applicability and proper analyzability of FMEA have caused to be among the most important techniques of systems for risk analysis and safety improvement. Risk management in all sectors is important, especially in agricultural sector. Sugarcane is one of the industrial crops used as raw material for several major and minor industries. In Iran, this crop is cultivated by sugarcane agro-industry companies. The sugarcane trailers were used to transport harvested sugarcane from farm to mill in these companies. There are many problems to milling it on time. One of the most important risks involved in sugarcane transportation is the delays encountered in this process which can affect the quality and quantity of the product. Delay in milling of the harvested sugarcane is caused by various reasons in agro-industry units including factory downtime, breakdowns of tractors at factory gate, tractor accident in factory yard and staff shift changes creating long queues. So, considering and using risk management techniques and eliminating risk factors can be an effective step to increase the efficiency of this process. Materials and Methods This research was carried out on Sugarcane and By-Products Development Company of Khuzestan. At first, the sugarcane transport operations and used equipment were investigated through an interview with experts in the safety and technical sectors and engineers of the Sugarcane and By-Products Development Company of Khuzestan and the study of related books in 2017. After that, the defects and errors of each equipment that caused technical problems and problems in other equipment, as well as the occurrence of injuries and human casualties were identified. Finally, the risks were written for valuation in the FMEA method paper. In this research, risk pricing was based on the Brainstorming method. Risk evaluation is based on the ranking of the effect severity, the risk occurrence probability and the degree of risk detection available in the FMEA method. In this research, analytical network process (ANP), a modern and powerful method in the decision-making field, has been used in combination with FMEA (FMEA-ANP) for defeating the shortcomings. FMEA-ANP considers mutual relationships of hazardous factors, and by offering a certain structure, develops a systematic and flexible view in risk management scope. The suggested method deploys a simple concept of risk priority number and assigns different importance in the form of power for each factor. The resulted RPN will cope better with the system, in which it is applied. This method provides a more accurate analysis of risk and, consequently, more efficient and effective actions, causing attainment and maintenance of more desirable reliability. Results and Discussion The results of FMEA-ANP model indicated that the mill equipment in the sugar factory is the most important delayed factor (failure) in the sugarcane transformation. For this reason, the basis failure causes in the sugar factory has been carefully investigated and it has been concluded by experts' opinions that factory mill and the conveyers failures are important causes of the delay in this process, respectively. Based on statistical analysis, 73.15% of the factory downtimes were related to mill and ranked as first compared with the other risk factors. Among the conveyors, the most damage was related to the inlet conveyor to the first mill and 49% of conveyors failures occurred in this conveyor. Conclusion This research validated the application of FMEA-ANP for the rational organization of the harvest-transport complex. According to this investigation, the probable downtimes and delays can be prevented by implementing the optimal preventive repair and maintenance planning in the sugar factory, and in particular on the factory mill equipment. In addition, efforts to adapt the speed of harvesting and the speed of delivery by the factory can be effective in reducing the delivery delay time by the factory.
