عنوان مقاله [English]
The main goal of this study was finding an appropriate pattern for estimating reliability of cereal harvester specially John Deer 955 combine in Farahan region. In this purpose, five steps including: selecting the statistical community, recording of data related to repairs and their abundance with the time of operation, choosing a suitable distribution function which fallow repairs, estimating the parameter of distribution function, and using estimated functions for determining of reliability were considered. In this case, five JD955 combine with 1800, 1000, 900, 800 and 445 worked hours were selected, randomly. Combines were divided to five main parts including: cutting and feeding unit, belts and chains, engine, hydraulic system and rest of combine mechanisms including: threshing system, separating system, cleaning system and transmition system. All of repairs were recorded based of time duration of work and systems. Repairs were divided into major and minor levels. Finally, time duration between impairments and their abundance types were recorded separately for each system. Based on data distribution and their variations, coefficient β>1, and general form of graph inferred from survival function, Weibull distribution function was selected as the best equation to make clear of impairments. The general form of Weibull function was selected with three parameters, but it changed to F(X)=1-EXP(-X^(β/α))form because first impairments could occurre in initial time of combine operation. In order to estimating of Weibull function parameters, maximum likelihood estimated (MLE) method was used. For anticipating the time between repairs, Mont carlo method was used. Results showed that reliability of studied combines was zero after 130 hours operation and time duration between repairs of combine was calculated by 42.8 hours.