An integrated systemic model for optimization of condition-based maintenance with human error
Introduction
Maintenance-related human errors have imposed heavy costs on industry. Research studies have reported on the significant role of maintenance-related human errors in aviation accidents [1], [2], [3], hazardous events in nuclear power plants [4], and software faults [5]. Dhillon and Liu [6] reported the impact of human errors in maintenance as found in the literature and come to the end with the finding that human error in maintenance is a pressing problem.
Generally, there exist two types of preventive maintenance schemes, i.e. time based maintenance (TBM) and condition-based maintenance (CBM). For CBM, the action taken after each inspection is dependent on the state of the system. It could be no action, or minimal maintenance to recover the system to the previous stage of degradation, or major maintenance to bring the system to as good as new state. For time-based preventive maintenance, the preventive maintenance is carried out at predetermined time intervals to bring the system to as good as new state.
Many researchers have reported on the superiority of condition based maintenance and predictive maintenance polices over traditional scheduled time based maintenance [7], [8]. However, in reality such strategies involve human in different functions including condition monitoring, diagnostics and prognostics. Human performance is not always perfect and therefore the effectiveness of new maintenance strategies should be assessed in presence of human errors.
The main objective of this study is to integrate human reliability model into the cost optimization of CBM. Moreover, the functional and organizational aspects of human error in CBM are studied. This paper proposes a model for CM scheduling optimization in which the concept of functional resonance is used to identify and report human errors in CBM. Specifically, human errors in CM including diagnostics and prognostics are separated from human errors in maintenance work including repair, replacement, and preventive maintenance. Both human errors in CM and maintenance are quantified in terms of human error probabilities (HEPs) and are embedded in the discrete-event simulation model of CBM system. Moreover, diagnostics and prognostics (D–P) algorithms are included in the system model with the aid of Monte Carlo simulation.
The contribution and significance of this study is fourfold. First, this is the first study that investigates human error in CBM functions and develops a systemic model for integration of human error in CBM optimization. Second, With reference to the functional and contextual analysis of human error in CBM, human error probabilities in different functions of CBM system are calculated. Third, this paper proposes an exact simulation-optimization model to optimize CM scheduling with respect to erroneous diagnostics and prognostics. To do this, the Monte Carlo simulation models of diagnostics and prognostics are embedded in the discrete-event simulation model of CBM system. This embedding enables optimization in a more realistic decision making environment. Fourth, a design of experiment is performed and the effects on optimum system cost of three factors: (i) human error in CM, (ii) human error in maintenance, and (iii) the accuracy and relevance of condition monitoring technology, are analyzed by the use of analysis of variance (ANOVA) technique.
The remainder of the paper is organized as follows. In Section 2 the current literature on modeling human error in maintenance and CBM optimization is reported. Section 3 presents an outline of the proposed systemic model and its subsequent 4 Human error in CBM system, 5 CBM system description elaborate the details of the proposed model components. Section 6 illustrates the simulation-optimization procedure used to solve the problem of CBM cost optimization. Section 7 presents an experiment with the proposed model and discussion on its results. Finally paper ends with main findings and conclusions.
Section snippets
CBM optimization
CBM optimization problem can be represented as a problem of finding some optimum decision variables for maintenance decision making based on a health criterion of the operating system. This health criterion could be the real degradation state of the system (x) or the monitored condition of the system (z). Table 1 summarizes research studies concerning CBM optimization. The main features of these studies are specified in three categories of CBM system specification, decision making, and problem
The integrated systemic model
Fig. 1 depicts an outlined representation of the proposed integrated systemic model for CBM cost optimization. In this presentation, ellipses and rectangular are representing parameters or variables of the CBM system while arrows represent computational methods or techniques integrated in this model. The model consists of two main modules, simulation and optimization. In simulation module, degradation models as well as diagnostics/prognostics algorithms are embedded in CBM system by the use of
Human error in CBM system
In some systems, there is a statutory requirement to investigate incidents and to assign a ‘cause’. This is the case, for example, with incidents involving aircraft. Frequently the cause assigned is human error. In fact, however, the error arises out of a quite specific combination of conditions in the man–machine system, including interface equipment, operational goals and procedures, and physical and social environments. It is preferable, therefore, to regard error as a function of the total
Degradation models
The general form of a degradation model can be described as a stochastic deterioration process with non-negative, stationary and independent increments. However, for the sake of being specific, two degradation models are selected from the existing literature, Fatigue crack degradation with Paris–Erdogan model [14] and Random positive normal shocks with Weibull frequency [16]. The descretized fatigue crack degradation with Paris–Erdogan model is:which represent a non-linear
Simulation
Based on what was mentioned, a discrete-event simulation procedure is developed and coded in MATLAB. Fig. 9 shows the pseudo-code of the simulation model. For a closer adherence to reality, the predictive model describing the evolution of the degrading systems as well as the performance of diagnostics/prognostics algorithms are embedded in the system model by the use of Monte Carlo (MC) simulation.
CBM cost optimization
The optimization model of CBM decides on the optimum values of RULc and α and minimizes the
Experiment design
In this section, to test the effects of contributing factors on AUC of the system, a complete balanced factorial design is conducted. With respect to degradation model, two distinct systems have been considered namely Paris–Erdogan and Random Shock. For both systems, the contributing factors and their levels are presented in Table 4. The factors HEPcm, HEPm, and are random because their levels are selected randomly.
Each of the combinations is considered with identical repair and PM times,
Conclusion
Human error in condition based maintenance is a multi-facet issue which calls for a systemic approach to be incorporated in maintenance optimization. This paper proposed an integrated systemic model for integration of human reliability model with condition based maintenance (CBM) optimization. With reference to the functional and contextual analysis of human error in CBM, human error probability in different functions of CBM system was studied.
The contribution and significance of this study is
Motivation and significance
Literature survey of this paper shows that there is a need to dedicate research works to development of models and techniques for investigation of human error in CBM optimization. This paper attempts to take a very first step in this area and develop a systemic model for incorporating human error in optimization of CBM system. Moreover, realistic operation management of CBM system decide CM scheduling based on the data derived from condition monitoring where the real health state of the
Acknowledgement
The authors are grateful for the valuable comments and suggestions by the respected reviewers, which have enhanced the strength and significance of this work. This study was supported by a grant from Iran National Science Foundation (Grant No. 92006268). The authors are grateful for the financial support provided by Iran National Science Foundation.
References (72)
- et al.
A framework for evaluating the effects of maintenance-related human errors in nuclear power plants
Reliab Eng Syst Saf
(2010) - et al.
Evaluating maintenance policies by quantitative modeling and analysis
Reliab Eng Syst Saf
(2013) - et al.
An overview of time-based and condition-based maintenance in industrial application
Comput Ind Eng
(2012) Stopping time optimization in condition monitoring
Reliab Eng Syst Saf
(2002)- et al.
Simulation modeling of non-repairable multi-component deteriorating systems for on condition maintenance optimization
Reliab Eng Syst Saf
(2002) - et al.
A condition-based maintenance policy for stochastically deteriorating systems
Reliab Eng Syst Saf
(2002) - et al.
A condition-based maintenance policy with non-periodic inspections for a two-unit series system
Reliab Eng Syst Saf
(2005) - et al.
Model-based Monte Carlo state estimation for condition-based component replacement
Reliab Eng Syst Saf
(2009) - et al.
A predictive maintenance policy with imperfect monitoring
Reliab Eng Syst Saf
(2010) - et al.
Discounted cost model for condition-based maintenance optimization
Reliab Eng Syst Saf
(2010)