XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

An enhanced approach for implementing Risk-Based Maintenance in a Total Productive Maintenance perspective

Lucantoni L.*, Bevilacqua M.*, Ciarapica F.E.*

*Dipartimento di Ingegneria Industriale e Scienze Matematiche, Università Politecnica Delle Marche, Via Brecce Bianche, 12 60131 - Ancona – Italy (l.lucantoni@staff.univpm.it, m.bevilacqua@staff.univpm.it,

f.ciarapica@staff.univpm.it)

Abstract: This paper discusses how to integrate Risk-Based Maintenance (RBM) methods into the Total Productive Maintenance (TPM) approach. The research approach outlined in this study is based on the description of a working procedure that learns from historical data and plans not only predictive but also proactive maintenance activities. Many researchers analysed the combined effects of different maintenance programs. However, topics such as RBM and TPM are generally analysed separately. This study investigates the practices of the two programs simultaneously because after successfully implementing RBM, TPM is needed for a correct equipment condition monitoring, maintenance improvement and planning too. In particular, the development of a single empirical framework can be useful as guidelines for diagnostics, fault prediction, and finally analysis in terms of losses and maintenance performance trends.

Keywords: Risk-Based Maintenance (RBM), Total Productive Maintenance (TPM), Predictive Maintenance (PdM)

1. Introduction

Over the last 70-odd years, maintenance approach has evolved from being essentially reactive (McKone & Weiss, 1998), to implementing preventive maintenance, to being condition-based and proactive. Today maintenance is one of the strategic business activities that increases company performance and its importance has changed over the years. In the eighties, with the automation of manufacturing companies, the organisational dimension plays an essential role in maintenance while today with globalisation and the need to reduce costs and time, the basic principles of maintenance are different for each company (Bevilacqua et al., 2018). Several companies adopted specific maintenance principles such as Risk-Based Maintenance (RBM) and Total Productive Maintenance (TPM) policies. RBM method defines maintenance priorities based on risk identification while TPM improves efficiency, quality, productivity, and some other indices. We develop a framework that integrates these maintenance policies to bridge their limits, improve the accuracy of data and information, and oprimise the management of maintenance activities and the availability of Big Data. The research gaps to be filled are identified using literature review. We observed that TPM methodologies neglect the aspects of risk analysis in the health, environmental and social fields, and the consequences and probabilities of occurrence of adverse events are not evaluated. On the other hand, RBM, usually conducted when a fault has already occurred, is a reactive process for maintenance decisions. However, it can be used to dynamically define the priority of intervention with specific algorithms relying on real-time data (Leonardi et al., 2019). Many companies are emphasising these shortcomings and need to combinate the two

methodologies. In the next section of this paper, we review RBM and TPM literature and present common practices of the two methods. Then in Sections 3 and 4, we treat our hypotheses, research approach and framework development. In Section 5, we discuss the results and analyse advantages and disadvantages of the integration. Finally, the conclusions are drawn.

2.State of the art

The literature review provides evidence of a renewed interest in the study of maintenance programs with an emphasis on their simultaneous investigation. However, there are no works about the integration of RBM and TPM techniques in the field of predictive maintenance. Since the 1980s many researchers examine the value of understanding the joint implementation of many manufacturing and maintenance programs conceptually, but a lot of methods are limited to their specific applications. McKone et al. (1999) propose a theoretical framework for understanding the use of TPM on managerial factors such as Just-In-Time (JIT), total quality management and employee involvement. Diamantoulaki & Angelides (2013) develop a framework for maintenance scheduling over the design life for an array of floating breakwaters connected with hinges. Sugumaran et al. (2014) investigate the integration of Quality Function Deployment (QFD) and Analytic Hierarchy Process (AHP) with TPM. McKone et al. (2001) investigate the relationship between TPM and manufacturing performance through Structural Equation Modeling, while Pascal et al. (2019) propose several maintenance managers indicators for reliability, diagnosis and prognosis to assess and improve the maintenance policy. Khan & Haddara (2003) still present a comprehensive and quantitative methodology for risk-based maintenance, while Jaderi et

53

XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

al. (2019) apply both traditional RBM and Fuzzy methods for the risk analysis of petrochemical assets failure. The development of dynamic RBM models using Bayesian Network is also interesting (Bhandari et al., 2016; Leoni et al., 2019). A review of the literature emphasises the use of probabilistic techniques for RBM studies, in fact, Duthie et al. (1998) integrate Probabilistic Assessment results into Reliability Centred Maintenance (RCM), that is part of the RBM approach, writing a Failure Mode Effects and Criticality Analysis (FMECA). Contrarily, TPM is a people-oriented concept that starts by fully harnessing the human intellectual capabilities which are normally hidden and unexploited (Ahuja & Khamba, 2008). About that, Vukadinovic et al. (2018) introduce the model of Early Human Resources Management in the Education and Training pillar of TPM, Bevilacqua & Ciarapica (2018) instead develop a procedure to integrate Human Factor in a refinery Risk Management System. Trapani et al. (2015) develop prognostics and health management systems in manufacturing. The entire literature on RBM and TPM is considered, and careful investigations into maintenance effectiveness have found that one third of its costs were wasted due to unnecessary or improper activities (Mobley, 2002). About that, data analysis becomes very important, and Gao et al. (2015) present a history of fault diagnosis using machine learning techniques. About that, while historical data is commonly used, collection and analysis of condition monitoring data have also been identified as

a useful approach in some of the existing RBM studies (Baliwangi et al., 2006; Omenzetter et al., 2016), in this paper too. Therefore, as previously mentioned, the paper contains a detailed review of the practices that are common to the RBM and TPM literature and, as shown in Table 1, our framework considers the relationship among the basic RBM and TPM techniques, and the preventive-oriented practices too. Next, we will present our hypothesised relationships drawing on these two theories.

2.1 Common practices to the two methods

The above literature review of RBM and TPM is summarised in Table 1. This analysis shows that the two programs include some common practices shared by both programs and other practices that are unique to each program. The idea of common practices has not been addressed in the literature. Nevertheless, RBM and TPM have different approaches but their similar practices include planned and preventive maintenance; information and feedback; equipment performance; probabilistic techniques; condition monitoring; Computerized Maintenance Management System (CMMS); safety, health, and environment. The last one is part of the TPM pillars, in fact, progress in the maintenance area has been motivated not only by the increase in number, size complexity and variety of physical resources; but also by the growing awareness of the maintenance impact on

Table 1: RBM-TPM practices or techniques commonly used

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31RBM Learning X X X X X X X X X X X X X X XRBM Identifying subsystems X X X X X X XRBM Hazard identification X X X X X X X X X X X X X X XRBM Likelihood assessment X X X X X XRBM Risk Estimation X X X X X X X X X XRBM Risk Assessment X X X X X X X X X X X X X X X XRBM Effects of failure X X X X X X X X X XRBM RCM X X X X X X X XRBM Quantitative Risk X X X X X X X X X XRBM Qualitative Risk X X X X X X XRBM Optimization X X X X X X X X X X X XRBM Loss estimation X X X X X X XRBM Re-evaluation risk X X XRBM Fault Tree Analysis X X X X XRBM FMECA X X XRBM Fuzzy RBM X XTPM Autonomous Maintenance X X X X X X X X X X X XTPM Quality Maintenance X X X X X X X XTPM Focused improvement X X X X X X X XTPM Employee involvement X X X X X X X X X X XTPM Equipment management X X X X X X X X X X X XTPM Education and Training X X X X X X X X X X X X X X XTPM Just In Time X X X XTPM Corrective maintenance X X XTPM Lean maintenance X X XCommon Planned maintenance X X X X X X X X X X X X X X X X X X X X X X X XCommon Preventive maintenance X X X X X XCommon Information and feedback X X X X X X X X X X X X XCommon Equipment performance X X X X X X X X X X X X X X X X X X X X XCommon Probabilistic techniques X X X X X X X X X X X X X X X XCommon Condition monitoring X X X X X X X X XCommon Safety, Health and Environoment X X X X X X X X X X X XCommon CMMs System X X X XCommon AHP X X X

RBM literaturea TPM literatureaMaintenance strategy

Practice or technique

a (1) Smith, 1989; (2) Duthie et al., 1998 (3) Stolen et al., 2002; (4) Khan & Haddara, 2003; (5) Krishnasamy et al., 2005; (6) Baliwangi et al., 2006; (7) Arunraj & Maiti, 2007; (8) Bertolini et al., 2009; (9) Clarke et al., 2010; (10) Diamantoulaki & Angelides, 2013; (11) Barone & Frangopol, 2014; (12) Mehairjan, 2017; (13) Cullum et al., 2018; (14) Jaderi et al., 2019; (15) Yazdi et al., 2019; (16) Al-Najjar, 1996; (17) McKone & Weiss, 1998; (18) McKone et al., 1999; (19) Chand & Shirvani, 2000; (20) Cua et al., 2001; (21) McKone et al., 2001; (22) Hansson et al., 2003; (23) Ahmed et al., 2004; (24) Rubrich & Watson, 2004; (25) Borris, 2006; (26) Singh et al., 2013; (27) Sugumaran et al., 2014; (28) Mwanza & Mbohwa, 2015; (29) Kiran, 2017; (30) Vukadinovic et al., 2018; (31) Pascal et al., 2019.

54

XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

the environment, personnel safety, company profitability and product quality (Khan & Haddara, 2003). Apart from the practices that are common to both programs, each of the methods also has unique practices that are more technical or process-oriented. RBM basic practices include for example hazard identification, fault tree analysis, and risk estimation and assessment. In particular, Khan & Haddara (2003) distinguished risk assessment in quantitative or qualitative. TPM basic practices are quality, autonomous and corrective maintenance, JIT, employee involvement, education and training, lean maintenance and focused improvement. Risk-based maintenance methodology provides a tool for maintenance planning and decision making to reduce the probability and consequences of equipment failure (Arunraj & Maiti, 2007). In the same way, Rimawan et al. (2018) affirm that TPM is a way to plan all maintenance activities include inspection and minor repairs to the stage overhaul. Therefore, it is stated that the main common point of the two methodologies is certainly the Planned Maintenance (PM) phase, as shown in Figure 1.

Figure 1: RBM and TPM intersection

3.Research approach

The research approach outlined in this study is a guideline for the development of self-learning platforms in the field of predictive and preventive maintenance of a production line, integrating two of the main maintenance approaches Risk-Based Maintenance and Total Productive Maintenance. According to McKone & Weiss (1998), the following represent the TPM general values:

▪ Process quality is a key part of every employee’s performance.

▪ Equipment failures can and will be prevented. ▪ If it is not broke, fix it anyway. ▪ Equipment performance can be managed.

The main framework objectives are the prevention of equipment failure and the schedule of maintenance activities, but the aim is also to respect the main targets emerging from the literature of both strategies: optimization of maintenance Key Performance Indicators (KPIs) (Chan et al., 2005; Montgomery & Serratella, 2002), knowledge of maintenance quality (Al-Najjar, 1996; Rubrich & Watson, 2004), improvement of Overall Equipment Effectiveness (OEE) (Rimawan et al., 2018), and monitoring of maintenance life cycle in terms of cost and time (Baliwangi et al., 2006; Barone & Frangopol, 2014). As shown in Figure 2, our framework consists of 5 modules: Learning, Diagnosis, Monitoring, Prognosis, Planning.

Figure 2: RBM and TPM strategy

The RBM approach is reflected more in the first two modules while TPM in the others. The third module marks the beginning of preventive maintenance procedures, and it will be activated only if the maintenance strategy is predictive. In this way, the principle of continuous improvement of TPM is respected without an excessive and unnecessary risk analysis. We hypothesise that applications, input/output data of each phase of the framework vary according to the case study. Each module consists of several steps, and each activity is an input/output system in which the information act as input for the next one. It is emphasised that the theoretical framework is based on Learning because often a crash is preceded by recurrent events. Moreover, the framework applicability requires a flexible Management Information System (MIS) to be used for decision-making. A Cloud computing system is needed for data storage and updating. It makes information technology system sources always available for analysis (Kwon et al., 2016). Since our hypotheses are not restrictive, our framework can be generalised.

4. Framework development

Figure 3 illustrates the proposed model and relationships for integrating of RBM and TPM methods. Moreover, three main streams of research provide support for our theoretical framework, i.e., risk theory, monitoring aspects, and proactive analysis. In general, the empirical framework purpose is to collect equipment history (Learning), analyse failure modes and maintenance policy (Diagnosis), control predictive signal by sensors (Monitoring), predict future (Prognosis) and finally, schedule and analyse maintenance activities (Planning). A detailed description of the methodology is presented in subsequent sections.

4.1 Module I: Learning

The development of a data warehouse is the starting point of our workflow model for the empirical examination of the maintenance system, as is. There is a variety of input data sources such as programs, users, external files, sensors for data logging, human operators, and more. Data acquisition and pre-processing phases need of Cloud databases and ETL operations (Extract, Transform and Load). About that, numerous tools exist to assist with the ETL process, such as DataStage, Informatica, or SQL and Server Integration Services (SSIS). In the case of production line, the Plant Data Layer consists of project data (i.e. layout, number and type of station, system classification, sensors, operational conditions),

55

XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

Figure 3: RBM and TPM integration framework: the proposed model and relationship

product data (i.e. input logic, bill of materials, production volume), process data (i.e. cycle time, processing time, recovery time, setup, Mean Time Between Failure o MTBF, Main Time to Repair o MTTR, reliability, OEE), technologies (i.e. software, sensors, information system), and historical data (i.e. equipment breakdowns, performance, maintenance techniques, interventions).The functional decomposition of a line considers the various stations, the individual machines, or even their components as a management unit. After gain better understanding of a process and identyfing the elementary parts by process mapping method (Robert Damelio, 2011), the proposed approach continues with the next module for each unit.

4.2 Module II: Diagnosis

Diagnosis is the process of detecting and identifying a failure mode within a management unit (what can happen, when, where, how, and why). It consists of the score attribution of severity, frequency, and detection capability for each failure. In this phase, we usually refer to archives of historical data on the different types of unwanted events. Components with a high Inspection Importance parameter contribute a large amount to the total plant risk and they must be identified and analysed (Duthie et al., 1998). Referring to the RBM, diagnostics is conducted to investigate the cause or nature of a problem from failure modes and risk identification to the attribution of maintenance policies. FMECA is one of the main techniques for the classification of unwanted events, and a Fault Tree Analysis is conducted to estimate the probability of failure for each of them. Instead, simulation techniques are used to explore the characteristics of the systems and to optimise their performances, estimating the consequences of a failure. Because risk can have

qualitative (expert judgment) or quantitative voices, according to Jaderi et al. (2019), the best thing is to develop a hybrid of both qualitative and quantitative analysis, using software like MATLAB. Then, if the Risk Priority Number (RPN) of the management unit is acceptable, the next unit is analysed; otherwise, the criticality matrix (Probability X Gravity) and root cause analysis are carried out. In this case, maintenance action is required to reduce the risk while the other cases (components with a low level of risk, high periodicity or low criticality) can be managed with relatively simple and periodic techniques, and even with corrective or event-driven actions. Maintenance policy selection is regarded as the cognitive process resulting in the selection of a belief or a course of action among several alternative possibilities including Preventive Maintenance, Predictive Maintenance (PdM), Breakdown Maintenance or Corrective Maintenance. The selection procedure can be implemented using simulation techniques (for example monitoring the OEE evolution based on maintenance activities), using the AHP method, or even both. In this way, the best maintenance policy is identified and associated with each unit of the critical analysis but, as specified in the paragraph 3.1., we can move to the next module only if the maintenance policy is PdM. It is the repair or replacement of machine components before failure, based on monitoring equipment operation, historical data, or predicted life cycles.

4.3 Module III: Monitoring

Monitoring is the module of the traditional diagnosis and implies activities such as obtaining measurable and quantified results and objectives, and continually monitoring and following through the process. This module is important to monitor the process degradation

56

XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

on which losses in performance and reliability of assets depend. Generally, degradation can be split into natural and forced degradation. Natural degradation is age or time-dependent. Forced degradation is external to systems. Several thresholds can be defined to provide a gradual degradation state for each component. Monitoring includes the definition of the possible sensors and the choice of the relative predictive signals according to the type of station, component, part, and other factors (e.g., pump vibrations, furnace temperature, wear of the gearboxes, etc.). The condition monitoring procedure is a comparison between the threshold value of the signal and the data collected and detected by the sensor in real time. If the real value of the signal exceeds the threshold value, but it is within the maintenance interval, a mathematical diagnosis is performed, and the process can move to the next module (Section 4.4). Otherwise, if the degradation level is outside the maintenance interval, failure is imminent, and an emergency repair is required.

4.4 Module IV: Prognosis

Prognosis generates future predictions, and mathematical and intelligent diagnosis must be developed. The first one includes the development of algorithms and the optimisations of maintenance KPIs to identify the correlations between data, parameterisation and fault pattern recognition, through mathematical models. The results of data analysis (procedures and indices) are the inputs for the intelligent diagnosis with the main purpose of predicting failure modes evolution. Intelligent diagnosis estimates the remaining useful life and remaining performance life until complete failure occurs. In this context, machine learning techniques are useful for the classification of data (assignment of observation to a class), regression (prediction of the numerical value of observation) and detection of forms. In this regard, we propose to use Neural Networks and Big Data (probability of failure over time) respectively for predictive and preventive maintenance, or even SVM method (Support Vector Machine). Maintenance KPIs optimisation treated maintenance scheduling as an optimisation problem because, according to Cullum et al. (2018), its key advantages over expert judgment are consistency and efficiency. The main KPIs are availability, which includes unplanned stops (such as breakdowns and other down events) and planned stops (such as changeovers); performance, which includes all factors that cause production to operate at less than the maximum possible speed when running (Slow Cycles and Small Stops); quality, which includes production rejects and reduced yield on start-up; and OEE, which includes all losses resulting in a measure of truly productive manufacturing time.

4.5 Module V: Planning

The prescriptive analysis which operates with the support of methodologies and tools useful to perceive in advance the problems, the trends or the future changes is useful to plan the opportune actions in time. Planning module schedules the predictive maintenance tasks based on measured and predicted failure rates. After evaluating

equipment and recording present status, restoring deterioration and improving weakness, building up information management system, and preparing predictive maintenance system by introducing equipment diagnostic techniques, maintenance scheduling is developed followed by estimation of costs and time for each maintenance activity. In particular, a life cycle can be predicted by the number of cycles, operation time, calendar time, component wear data, variations in the operating parameters of the component such as temperature or vibration (Rubrich & Watson, 2004). After the application of the maintenance plan, the management system should send feedback from the operational procedure to update process performance and implement a continuous improvement system. “Information and feedback” consists of data storage and analysis update, which includes parameters such as MTTR, MTBF, RPN, OEE, and CAE (Corrective Action Effectiveness).

5.Discussion

The RBM-TPM procedure carried out in this paper is a dynamic and iterative process. In the previous sections, we used a table and two figures to identify the common practices to both techniques and to highlight the most critical practice among these: planning maintenance. Our framework emphasises a procedure to implement RBM as a preliminary phase of TPM philosophy, to improve maintenance effectiveness and planning. The main goal of this research is to develop a common framework useful to integrate two methodologies with different theories and techniques. If the company has the goal of cost reduction, TPM practices and techniques can be the best ones (Cua et al., 2001). On another hand, if the maintenance activities aim at high quality and reliability, practices like RCM or RBM, might be the best. The common problems in the implementation of both maintenance programs are related to human- and socio-technical systems theory. The key difference between TPM and RBM is that the first one requires the involvement of all the people in the organisation (Rubrich & Watson, 2004), while TPM is based on the "management of motivation" and involves the activities of small groups and single employees (Chan et al., 2005). RBM does not focus on these aspects, and this is certainly a significant distance between the two methodologies. It is possible to overcome this problem with a framework that develops towards aspects such as learning operators to perform important daily maintenance tasks, to identify and resolve many basic equipment problems, to reduce human errors and to improve human performance in each pre- and post-maintenance task. About that, the framework implementation must pay attention to possible frequent and erroneous problem repetition, traceability/visibility of maintenance program, blind acceptance of input information, difficulty in data management, and possible high implementation costs. It is important to keep in mind that the realised framework is universal but, according to the needs, requires some modifications and adaptations. It is clear from the RBM and TPM integration that our framework is not limited to random failures, on the contrary, proposes condition monitoring and data analysis.

57

XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

One of the advantages of the framework is certainly the link established between two maintenance strategies that act differently. In this way, it is possible to assimilate the advantages of both techniques. The integration is useful to perform just-in-time maintenance, avoid failure, minimise equipment downtime and, with small changes, also to ensure product quality. Another aspect is that RBM and TPM integration can improve data accuracy providing information in real-time (which instead are normally uncertain and inaccurate). The integration can provide maintenance planners with the information necessary for the development of predictive maintenance programs and more effective control of activities, and it can offer management a global vision of the health status of the equipment. About that, new human digital assistant systems can contribute to overcome the weaknesses of human- and socio-technical systems differently depending on the case. An adeguate real time information system about all RBM factors is an intuitive employee guidance and can help maintenance managers in making faster objective decisions, based on more reliable data. In the case of TPM, could be interesting learning about work satisfaction with elements of artificial intelligence (i.g. smart clothing, smart bracelets), in order to analyse ergonomic aspects and to obtaine a more efficient workforce. Finally, in the RBM and TPM combination, digital and assistant systems contribute to the integration of human- and socio-technical systems in dynamic relationship with their environment. Augmented and virtual reality connect virtual information to the physical world, and continuous networking supports mobile service technicians 4.0 in their different tasks: from preparing for maintenance operations to reporting the results.

6. Conclusion

The primary purpose of this paper is to point out the possibility to improve the ways data and information are treated in Total Production Maintenance using the Risk-Based Maintenance principle. The authors introduced a model designed through the integration of procedures of pre-posterior analysis and continuous improvement, respectively the Impact Evaluation of RBM and Planned Maintenance of TPM. The paper tries to answer the question "how to have an overview of two seemingly distant strategies?". About that, the research illustrates the theoretical process for turning data in information, integrating Lean philosophy with concepts such as maintenance reliability, risk and planning, and then developing a single empirical framework. Our research emphasises that maintenance planning is the main pillar of both RBM and TPM techniques and, summing up, the working procedure involves analysing system data, classifying risks, defining the best maintenance strategy and identifying the main tasks for correct predictive maintenance. With the aim to improve maintenance performance and to reduce losses, maintenance policy should be transformed from corrective to proactive and planned too. The proposed framework provides a theoretical means for better understanding equipment performance, the history of failure events, what common problems can occur and how those problems can be prevented through early detection and treatment of

abnormal conditions. Integration is not immediate, and the variants are multiple. This result is consistent with the initial expectation not only to integrate the two strategies but to extend them with proactive activities, going to predict costs, faults and future performance trends.

References

Ahmed, S., Hassan, M. H., & Taha, Z. (2004). State of implementation of TPM in SMIs: A survey study in Malaysia. Journal of Quality in Maintenance Engineering, 10(2), 93–106.

Ahuja, I. P. S., & Khamba, J. S. (2008). Total productive maintenance implementation in a manufacturing organisation. International Journal of Productivity and Quality Management, 3(3), 360.

Al-Najjar, B. (1996). Total quality maintenance: An approach for continuous reduction in costs of quality products. Journal of Quality in Maintenance Engineering, 2(3), 4–20.

Arunraj, N. S., & Maiti, J. (2007). Risk-based maintenance-Techniques and applications. Journal of Hazardous Materials, 142, 653–661.

Baliwangi, L., Ishida, K., Arima, H., & Artana, K. B. (2006). Optimizing Ship Machinery Maintenance Scheduling Through Risk Analysis and Life Cycle Cost Analysis. 25th International Conference on Offshore Mechanics and Arctic Engineering, 127–134.

Barone, G., & Frangopol, D. M. (2014). Reliability, risk and lifetime distributions as performance indicators for life-cycle maintenance of deteriorating structures. Reliability Engineering and System Safety, 123, 21–37.

Bertolini, M., Bevilacqua, M., Ciarapica, F. E., & Giacchetta, G. (2009). Development of Risk-Based Inspection and Maintenance procedures for an oil refinery. Journal of Loss Prevention in the Process Industries, 22(2), 244–253.

Bevilacqua, M., Ciarapica, F. E., & Mazzuto, G. (2018). Fuzzy cognitive maps for adverse drug event risk management. Safety Science.

Bevilacqua, Maurizio, & Ciarapica, F. E. (2018). Human factor risk management in the process industry: A case study. Reliability Engineering and System Safety, 169, 149–159.

Bhandari, J., Arzaghi, E., Abbassi, R., Garaniya, V., & Khan, F. (2016). Dynamic Risk-Based Maintenance for Offshore Processing Facility, 35(4), 399–406.

Borris, S. (2006). TPM Proven stratagies and techiques to keep equipment runnig at peak efficciency. New York: McGraw-Hill.

Chan, F. T. S., Lau, H. C. W., Ip, R. W. L., Chan, H. K., & Kong, S. (2005). Implementation of total productive maintenance: A case study. International Journal of Production Economics.

Chand, G., & Shirvani, B. (2000). Implementation of TPM in cellular manufacture. Journal of Materials Processing Technology, 103, 149–154.

Clarke, G., Mulryan, G., & Liggan, P. (2010). Lean Maintenance – A Risk-Based Approach. International Society for Pharmaceutical Engineering.

Cua, K., McKone, K., & Schroeder, R. (2001). Relationships between implementation of TQM,

58

XXIV Summer School “Francesco Turco” – Industrial Systems Engineering

JIT, and TPM and manufacturing performance. Journal of Operations Management, 19(6), 675–694.

Cullum, J., Binns, J., Lonsdale, M., Abbassi, R., & Garaniya, V. (2018). Risk-Based Maintenance Scheduling with application to naval vessels and ships. Ocean Engineering, 148, 476–485. 4

Diamantoulaki, I., & Angelides, D. C. (2013). Risk-based maintenance scheduling using monitoring data for moored floating breakwaters. Structural Safety, 41, 107–118.

Duthie, J., Robertson, M., Clayton, A., & Lidbury, D. (1998). Risk-based approaches to ageing and maintenance management. Nuclear Engineering and Design, 184, 27–38.

Gao, Z., Cecati, C., & Ding, S. X. (2015). A Survey of Fault Diagnosis and Fault-Tolerant Techniques Part I: Fault Diagnosis. Industrial Electronics, IEEE Transactions On.

Hansson, J., Backlund, F., & Lycke, L. (2003). Managing commitment: Increasing the odds for successful implementation of TQM, TPM or RCM. International Journal of Quality & Reliability Management, 20(9), 993–1008.

Jaderi, F., Ibrahim, Z. Z., & Zahiri, M. R. (2019). Criticality analysis of petrochemical assets using risk based maintenance and the fuzzy inference system. Process Safety and Environmental Protection, 121, 312–325.

Khan, F. I., & Haddara, M. M. (2003). Risk-based maintenance (RBM): A quantitative approach for maintenance/inspection scheduling and planning. Journal of Loss Prevention in the Process Industries, 16(6), 561–573.

Kiran, D. R. (2017). Chapter 13 – Total Productive Maintenance. In Total Quality Management (pp. 177–192).

Krishnasamy, L., Khan, F., & Haddara, M. (2005). Development of a risk-based maintenance (RBM) strategy for a power-generating plant. Journal of Loss Prevention in the Process Industries, 18, 69–81.

Kwon, D., Hodkiewicz, M. R., Fan, J., Shibutani, T., & Pecht, M. G. (2016). IoT-Based Prognostics and Systems Health Management for Industrial

Leonardi, F., Messina, F., & Santoro, C. (2019). A Risk-Based Approach to Automate Preventive Maintenance Tasks Generation by Exploiting Autonomous Robot Inspections in Wind Farms.

Leoni, L., BahooToroody, A., De Carlo, F., & Paltrinieri, N. (2019). Developing a risk-based maintenance model for a Natural Gas Regulating and Metering Station using Bayesian Network. Journal of Loss Prevention in the Process Industries, 57(October 2018), 17–24.

McKone, Kathleen E., Schroeder, R. G., & Cua, K. O. (1999). Total productive maintenance: A contextual view. Journal of Operations Management, 17, 123–144.

McKone, Kathleen E., Schroeder, R. G., & Cua, K. O. (2001). Impact of total productive maintenance practices on manufacturing performance. Journal of Operations Management, 19(1), 39–58.

McKone, KATHLEEN E., & Weiss, E. N. (1998). TPM: Planned and autonomous maintenance: bridging the

gap between practice and research. Production and Operations Management, 7, 335–351.

Mehairjan, R. P. Y. (2017). Risk-Based Maintenance for Electricity Network Organizations. Risk-Based Maintenance for Electricity Network Organizations, 9–30.

Mobley, R. K. (2002). An Introduction to Predictive Maintenance (Second Edition). USA: Elsevier Science.

Montgomery, R.L., Serratella, C. (2002). Risk-based maintenance: a new vision for asset integrity management. ASME 2002 Pressure Vessels and Piping Conference, 151–165.

Mwanza, B. G., & Mbohwa, C. (2015). Design of a Total Productive Maintenance Model for Effective Implementation: Case Study of a Chemical Manufacturing Company. Procedia Manufacturing, 4(Iess), 461–470.

Pascal, V., Toufik, A., Manuel, A., Florent, D., & Frédéric, K. (2019). Improvement indicators for Total Productive Maintenance policy. Control Engineering Practice, 82, 86–96.

Rimawan, E., Kholil, M., & Hendri, H. (2018). Measurement value analysis overall equipment effectiveness (OEE) packaging process in line 2 (Case Study of PT. MBI Tbk). In IOP Conference Series: Materials Science and Engineering.

Robert Damelio. (2011). The Basics of Process Mapping (2nd Editio). New York.

Rubrich, L., & Watson, M. (2004). Implementing World Class Manufacturing (Second Edi). Fort Wayne, Indiana: WCM Associates.

Singh, R., Gohil, A. M., Shah, D. B., & Desai, S. (2013). Total productive maintenance (TPM) implementation in a machine shop: A case study. Procedia Engineering, 51, 592–599.

Smith, B. D. (1989). Risk Management in Repair Work Decision Making. Naval Engineers Journal, 220–230.

Stolen, K., Braber, F., Dimitrakos, T., Fredriksen, R., Gran, B., Houmb, S., … Aagedal, J. (2002). Model-based risk assessment-the CORAS approach. 1st ITrust Workshop (2002), (January 2001).

Sugumaran, C., Muthu, S., Devadasan, S. R., Srinivasan, K., Sivaram, N. M., & Rupavathi, N. (2014). Integration of QFD and AHP with TPM: an implementation study in an automotive accessories manufacturing company. International Journal of Productivity and Quality Management, 14(3), 263.

Trapani, N., Fumagalli, L., & Macchi, M. (2015). Risk driven engineering of Prognostics and Health Management systems in manufacturing. IFAC-PapersOnLine, 48(3), 995–1000.

Vukadinovic, S., Macuzic, I., Djapan, M., & Milosevic, M. (2018). Early management of human factors in lean industrial systems. Safety Science, 1–7.

Yazdi, M., Nedjati, A., & Abbassi, R. (2019). Fuzzy dynamic risk-based maintenance investment optimization for offshore process facilities. Journal of Loss Prevention in the Process Industries, 57, 194–207.

59