Prioritization and Evaluation of Mechanical Components Failure of CNC Lathe Machine based on Fuzzy FMEA Approach
Journal Title: Journal of Agricultural Machinery - Year 2019, Vol 9, Issue 2
Abstract
Introduction In recent years, with development of industrial products with complex and precise systems, the demand for CNC machines has been increasing, and as its technology has been progressed, more failure modes have been developed with complex and multi-purpose structures. The necessity of CNC machinesâ reliability is also more evident than ever due to its impact on production and its implementation costs. Aiming at reducing the risks and managing the performance of the CNC machine parts in order to increase the reliability and reduce the stop time, it is important to identify all of the failure modes and prioritize them to determine the critical modes and take the proper cautionary maintenance actions approach. Materials and Methods     In this study, conventional and fuzzy FMEA, which is a method in the field of reliability applications, was used to determine the risks in mechanical components of CNC lathe machine and all its potential failure modes. The extracted information was mainly obtained by asking from CNC machine experts and analysts, who provided detailed information about the CNC machining process. These experts used linguistic terms to prioritize the S, O and D parameters. In the conventional method, the RPN numbers were calculated and prioritized for different subsystems. Then in the fuzzy method, first the working process of the CNC machine and the mechanism of its components were studied. Also, in this step, all failure modes of mechanical components of the CNC and their effects were determined. Subsequently, each of the three parameters S, O, and D were evaluated for each of the failure modes and their rankings. For ranking using the crisp data, usually, the numbers in 1-10 scale are used, then using linguistic variables, the crisp values are converted into fuzzy values (fuzzification). 125 rules were used to control the output values for correcting the input parameters (Inference). For converting input parameters to fuzzy values and transferring qualitative rules into quantitative results, Fuzzy Mamdani Inference Algorithm was used (Inference). In the following, the inference output values are converted into non-fuzzy values (defuzzification). In the end, the fuzzy RPNs calculated by the fuzzy algorithm and defuzzified are ranked. Results and Discussion In conventional FMEA method, after calculating the RPNs and prioritizing them, the results showed that this method grouped 30 subsystems into 30 risk groups due to the RPN equalization of some subsystems, while it is evident that by changing the subsystem, the nature of its failure and its severity would vary. Therefore, this result is not consistent with reality. According to the weaknesses of this method, fuzzy logic was used for better prioritization. In the fuzzy method, the results showed that, in the 5-point scale, with the Gaussian membership function and the Centroid defuzzification method, it was able to prioritize subsystems in 30 risk groups. In this method, gearboxes, linear guideway, and fittings had the highest priority in terms of the criticality of failure, respectively. Conclusion The results of the fuzzy FMEA method showed that, among the mechanical systems of CNC lathe machine, the axes components and the lubrication system have the highest FRPNs and degree of criticality, respectively. Using the fuzzy FMEA method, the experts' problems in prioritizing critical modes were solved. In fact, using the linguistic variables enabled experts to have a more realistic judgment of CNC machine components, and thus, compared to the conventional method, the results of the prioritization of failure modes are more accurate, realistic and sensible. Also, using this method, the limitations of the conventional method were reduced, and failure modes were prioritized more effectively and efficiently. Fuzzy FMEA is found to be an effective tool for prioritizing critical failure modes of mechanical components in CNC lathe machines. The results can also be used in arranging maintenance schedule to take corrective measures, and thereby, it can increase the reliability of the machining process.
Authors and Affiliations
A. Vaysi,A. Rohani,M. Tabasizadeh,R. Khodabakhshian,F. Kolahan,
Keywords
Related Articles
- EP ID EP717934
- DOI https://doi.org/10.22067/jam.v9i2.73103
- Views 62
- Downloads 0
Measurement of Losses in a Austoft Sugarcane Harvester Case 7000
IntroductionSugarcane is one of the strategic products of Khuzestan province, which is cultivated in 10 active agro-industrial sites and covers an area of about 110,000 hectares of irrigated farms in the province. Sugarc...
Evaluating the Energy Consumption and Environmental Impacts in Milk Production Chain (Case Study: Kermanshah City of Iran)
Introduction Life cycle assessment of food products is an appropriate method to understand the energy consumption and production of environmental burdens. Dairy production process has considerable effect on climate chang...
Spectral Feature Selection from the Hyperspectral Dataset to Identify Pistachio Leaves Infected by Psylla
IntroductionPistachio production has been adversely affected by Psylla, which is a devastating insect. The primary goal of this study was to select sensitive spectral bands to distinguish pistachio leaves infected by Psy...
Measuring Mechanical Properties and Simulating the Creep Behavior of Wild Cane Stem using Burger-Voigt Model
IntroductionKnowledge of mechanical and viscoelastic properties of agricultural material will be helpful in the transportation and processing of these materials. Giant reed grass, also known as wild cane, is a tall, pere...
Design and Construction of a High-voltage System to Kill Weeds with a Feedback Mechanism
IntroductionVarious methods have been performed to control weeds in the world and the use of herbicides is one of them, but public concerns about human health have changed interest in alternative methods. Thermal methods...