2nd National Conference on Latest Develeopments in Materials, Manufacturing and Quality Control from 13-14th February, 2014.

Maintenance criticality analysis using TOPSIS

Anish Sachdeva1, Pradeep Kumar2, Dinesh Kumar2
1

Department of Industrial and Production Engineering, National Institute of Technology, Jalandhar, India (anishsachdeva@gmail.com) 2 Department of Mechanical and Industrial Engineering, Indian Institute of Technology, Roorkee, India

Abstract - In this paper, a multi factor decision making approach is presented for prioritizing failures causes for the digester of pulping system of a paper mill as an alternative to traditional approach of failure mode and effect analysis (FMEA). The approach is based on the technique for order preference by similarity to ideal solution (TOPSIS). The priority ranking is formulated on the basis of six parameters (failure occurrence, non detection, maintainability, spare parts, economic safety and economic cost). The Shannon's entropy concept is used for assigning objective weights to maintenance parameters. Keywords - Maintenance, TOPSIS, FMECA, and

MCDM, Risk priority number. I. INTRODUCTION Deciding on the best maintenance policy is not an easy matter as the maintenance program must combine technical requirements with the management strategy. A good maintenance program must define maintenance strategies for different facilities. The failure mode of every component must be studied in order to assess the best maintenance solution, in accordance with its failure pattern, impact and cost on the whole system. This information helps the maintenance personnel to decide the best suited maintenance action and to assign the different priorities to various plant components and machines. The management of large number of tangible and intangible attributes that must be taken into account represents the complexity of the problem. Several techniques have been discussed in the literature for planning maintenance activities of industrial plants. The most commonly used technique to evaluate the maintenance significance of the items/ failure modes and categorise these in several groups of risk is based on using failure mode effect and criticality analysis FMCEA. This methodology has been proposed in different possible variants, in terms of relevant criteria considered and/or risk priority number formulation. Using this approach, the selection of a maintenance policy is performed through the analysis of obtained priority risk number. This technique was first proposed by NASA in 1963 for their obvious reliability requirements. Since then, it has been extensively used as a powerful technique for system safety and reliability analysis of products and processes in a wide range of industries [1], [2]. The main objective of FMEA is to discover and prioritise the potential failure modes by computing respective RPN, which is a product of failure occurrence, severity and

probability of non detection of a fault. Though RPN evaluation with FMEA is probably the most popular technique for reliability and failure mode analysis, several problems are associated with its practical implementation, which have been addressed by many authors [3],[4],[5],[6],[7]. The most important problems discussed were regarding the - FMCEA does not consider the interdependence among the various failure modes and effects; - It considers only three kinds of attributes whereas other important aspects like economic aspects, production quantities, and safety aspects etc. are not taken into consideration - It is assumed that the three indexes are equally important and identifying situations with the same priority number characterized by different index levels. - the assumption that the scales of three severity (S), occurrence (O) and detection (D) indexes have the same metric and that the same design level corresponds to the same values on different index scales; - different sets of the three factors can produce exactly the same value of RPN, but the hidden implication may be totally different - The method of multiplication adopted for calculating the risk priority number is questionable. Considering the importance and complexity of the maintenance design problem it is observed that further efforts are needed for the development of effective methods which will incorporate numerous evaluation criteria and help the maintenance staff in evaluating the impact of intangible factors in the maintenance decision making and identify the best maintenance policy accordingly. In this paper, a new technique based on modified FMEA alongwith TOPSIS is proposed to determine the risk priority number and to overcome the limitations of the conventional RPN, as cited above. This technique permits to take into consideration the several possible aspects concerning the maintenance selection problem (failure chances, detectability, costs and safety aspect etc.) The method is based on a technique for order preference by similarity to ideal solution (TOPSIS). TOPSIS is a multiattribute decision making methodology based on the measurement of the Euclidean distance of an alternative from an ideal goal [8]. The procedure for TOPSIS methodology is presented in the subsequent section.

Occurrence Almost never Rare Very few few Medium Moderately high High Very high Extremely high

TABLE 1 SCORE FOR CHANCE OF FAILURE MTBF Score >3 years 2-3 years 1-2 years 3/4-1 year 6 - 9 months 4 - 6 months 2 - 4 months 1 - 2 months < 30 days 1 2 3 4 5 6 7 8 9

TABLE 5 SCORES FOR ECONOMIC SAFETY LOSS Status of the equipment/ sub system With no moving parts With one moving part/ critical category With two moving parts/ critical category With three moving parts/ critical category With more than three moving parts/ critical category TABLE 6 SCORES FOR ECONOMIC COST Criteria for Economic Cost Extremely low Very low Low Fair Medium Moderately high High Very high Extremely high

Score 2 3 5 7 9

Score 1 2 3 4 5 6 7 8 9

TABLE 2 SCORE FOR NON DETECTION OF FAILURES Likelihood of Criteria for non detection of Score Non-detection (%) failures < 10 Extremely low 1 10-20 Very low 2 21-30 31-40 41-50 51-60 61-70 71-80 > 80 Low fair medium moderately high High Very high Extremely high 3 4 5 6 7 8 9

TABLE 7 SCORES FOR FAILURE CAUSES OF DIGESTER

Major Components and Potential cause of failure Wire mesh Abrasion of mesh [D1] Corrosion [D2] Foreign material [D3] Vacuum pumps Lack of lubrication in moving parts. [D4 ] Bearing failure [D5] Inclusion of solid particles[D6] Seal failure [D7] Let down relief valve Mechanical failure [D8] Blockage. [D9] O 3 5 8 D 5 5 8 M 3 6 4 SP 4 4 2 ES 3 3 3 EC 3 4 4

Criteria Mt > 0.8 0.7 < Mt 0.8 0.6 < Mt 0.7 0.5 < Mt 0.6 0.4 < Mt 0.5 0.3 < Mt 0.4 0.2 < Mt 0.3 0.1 < Mt 0.2 Mt < 0.1

TABLE 3 SCORE FOR MAINTAINABILITY Maintainability Almost certain Very high High Moderately high Medium Low Very low Slight Extremely low

Score 1 2 3 4 5 6 7 8 9

4 5 5 8

5 5 7 5

6 8 5 6

4 4 3 3

6 6 6 3

5 7 7 5

5 4

8 7

5 4

8 7

3 3

7 6

Figure labels should be legible, approximately 8 to 10 point type.

TABLE 4 SCORING CRITERIA FOR SPARE PARTS Availability Criticality Easy Difficult Scarce Desirable 1 4 7 Essential 2 5 8 Vital 3 6 9

Fig. 1 Mapping nonlinear data to a higher dimensional feature space

2nd National Conference on Latest Develeopments in Materials, Manufacturing and Quality Control from 13-14th February, 2014.
V. CONCLUSIONS This paper presents a new modified FMECA approach to deal with the problems encountered while defining the best mix of maintenance policies. An objective weighted function based multi criteria failure mode analysis technique using TOPSIS is proposed to find more accurate and reliable priority risk numbers for performing the criticality analysis. This enables to obtain a ranking of failure modes/components by incorporating several types of information related to performance, safety and society. In particular, the analysis of prioritization of failure causes provides a framework to decide upon the type of maintenance strategies for different failure modes. If reliable quantitative judgments are available for some criteria, then it can also be easily included in the analysis. So the use of the proposed approach forms a basis for the continuous process of reliability design and maintenance strategy decisions.
[1] [2]

REFERENCES
P. O'Connor, Practical Reliability Engineering, Singapore: John Wiley & Sons, 2003. S. H. Teng, and S. Y. Ho, Failure mode and effects analysis: an integrated approach for product design and process control , International Journal of Quality & Reliability Management , vol. 13, no. 5, pp. 8-26, 1996. B. G. Dale and C. Cooper, Total Quality and Human Resources: An Executive Guide, Oxford :Blackwell Publishers, 1992. D. Straker, A Tool book for Quality Importance and Problem Solving, London: Prentice-Hall International Limited, 1995. J. B. Bowles and C. E. Pelaez, Fuzzy logic prioritization of failures in a system failure mode, effects and criticality analysis, Reliability Engineering and System Safety, vol. 50, pp. 203213, 1995. M. Braglia, MAFMA: multi-attribute failure mode analysis, International Journal of Quality & Reliability Management , vol. 17, pp. 1017-1033, 2000. N. R. Shankar and B. S. Prabhu, Modified approach for prioritization of failures in a system failure mode and effects analysis , International Journal of Quality & Reliability Management, vol. 18, no. 3, pp. 324-335, 2001. M. Braglia, M. Frosolini, and R. Montanari, Fuzzy TOPSIS approach for failure mode, effects and criticality analysis, Quality and Reliability Engineering International, vol. 19, pp. 425443, 2003. C. L. Hwang, and K. Yoon, Multi Attribute Decision Making Methods and Applications, New York: Springer-Verlag, 1981. C. Parkan, and M. Wu, Decision-making and performance measurement models with applications to robot selection, Computers & Industrial Engineering, vol. 36, pp. 503-23,1999. D. Jee, and K. Kang, A method for optimal material selection aided with decision making theory, Materials and Design, vol. 21, pp. 199-206, 2000. H. Deng, C. Yeh and R. J. Willis, Inter-company comparison using modified TOPSIS with objective weights, Computers & Operations Research, vol. 27, pp. 963-973, 2000.

[3] [4] [5]

[6]

[7]

[8]

[9] [10]

[11]

[12]

APPPENDIX A THE DISTANCES OF FAILURE CAUSES FROM IDEAL SOLUTION O Failure Causes D1 D2 D3 D4 D5 D6 D7 D8 D9 D M SP ES EC

+ i1

i1

+ i2

i2

+ i3

i3

+ i4

i4

+ i5

i5

+ i6

d i6

0.1064 0.0638 0.0000 0.0851 0.0638 0.0638 0.0000 0.0638 0.0851

0.0000 0.0426 0.1064 0.0213 0.0426 0.0426 0.1064 0.0426 0.0213

0.0545 0.0545 0.0000 0.0545 0.0545 0.0182 0.0545 0.0000 0.0182

0.0000 0.0000 0.0545 0.0000 0.0000 0.0364 0.0000 0.0545 0.0364

0.1064 0.0426 0.0851 0.0426 0.0000 0.0638 0.0426 0.0638 0.0851

0.0000 0.0638 0.0213 0.0638 0.1064 0.0426 0.0638 0.0426 0.0213

0.0571 0.0571 0.1143 0.0571 0.0571 0.0857 0.0857 0.0000 0.0286

0.0571 0.0571 0.0000 0.0571 0.0571 0.0286 0.0286 0.1143 0.0857

0.0476 0.0476 0.0476 0.0000 0.0000 0.0000 0.0476 0.0476 0.0476

0.0000 0.0000 0.0000 0.0476 0.0476 0.0476 0.0000 0.0000 0.0000

0.0833 0.0625 0.0625 0.0417 0.0000 0.0000 0.0417 0.0000 0.0208

0.0000 0.0208 0.0208 0.0417 0.0833 0.0833 0.0417 0.0833 0.0625

The following is a summary of the formatting used and to be followed during the submission of your paper: Font: Times New Roman Font Size: v Title 14 pt v Authors 12 pt v Authors Affiliation 10 pt v Abstract 9 pt bold v Body Text 10 pt v Table Text 8 pt v Figure Labels 8-10 pt v References Text 8 pt Margins: v Top 0.98 v Left 0.75 v Bottom 1.22 v Right 0.75 v Gutter 0 v Gutter Position Left Orientation: Portrait Paper: A4 Width 8.27 Height 11.69 Layout: Headers and Footers: As shown in the Sample Paper From Edge: Header 0.5 Footer 0.5