Off-line Deep Reinforcement Learning for Maintenance Optimization

Several machine learning and deep learning frameworks have been proposed to solve remaining useful life estimation and failure prediction problems in recent years. Having access to the remaining useful estimation or the likelihood of failure in the near future help operators to assess the operating conditions and, therefore, making better repair and maintenance decisions. However, many operators believe remaining useful life estimation and failure prediction solutions are incomplete answers to the maintenance challenge. They would argue that knowing the likelihood of failure in a given time interval or having access to an estimation of the remaining useful life are not enough to make maintenance decisions which minimize the cost while keeping them safe. In this paper, we present a maintenance framework based on off-line deep reinforcement learning which instead of providing information such as likelihood of failure, suggests actions such as “continue the operation” or “visit a repair shop” to the operators in order to maximize the overall profit. Using off-line reinforcement learning makes it possible to learn the optimum maintenance policy from historical data without relying on expensive simulators. We demonstrate the application of our solution in a case study using NASA C-MAPSS dataset.

Multi-objective maintenance planning under preventive maintenance

PurposeThis research developed and examined a mathematical model for concurrent corrective and preventive maintenance policy of a system of series configuration.Design/methodology/approachA mathematical model was developed to maximize availability, and maximal net revenues, and minimal cost. Different probability distributions for time to failure and time to repair were considered. The model was then implemented on a real case study, which was studied under corrective maintenance policy and concurrent corrective and preventive policy.FindingsA comparison between results at current policy (90 days) and optimal period of corrective and preventive policy was conducted. It was found that availability, profit was increased from 94.4% and $20.091 – 96.5% and $24.803, respectively. Further, the cost was reduced from $1104.8 to $797.22.Research limitations/implicationsThe proposed optimization model can be adopted in planning maintenance activities for a single machine as well as for a system of series configuration machines under various probability distributions.Practical implicationsThe proposed model can significantly enhance performance of the production as well as maintenance systems. In addition, the developed model may support maintenance engineering in effective management of maintenance resources and the performance of its activities.Originality/valueThis research considers a mathematical model with multi-objective functions and distinct probability distributions for time-to-failure for a system of series machines. Moreover, appropriate approximation solution was deployed to find integral of some functions. Finally, it provides maintenance planning for a single machine or a series of machines.

Opportunistic perfect preventive maintenance policy in presence of masked data

In this paper, the maintenance optimization problem of multi-component system is considered. It is assumed that the exact cause of system failure might be masked. That is, the exact cause of failure is unknown, and we only know that it belongs to a set called mask set. Both opportunistic perfect preventive maintenance (OPPM) and perfect corrective maintenance are considered. Threshold of OPPM and inter-inspection interval are considered as decision parameters which are optimized using long-run cost rate criteria. The applicability of the proposed maintenance policy is investigated using an illustrative example.

A Model for Managing the Performance of CMMS Deployment in High-Rise Office Buildings: A View from Lagos, Nigeria

The development of high-rise buildings is a current trend in developed cities to answer the challenges of population growth, adding aesthetic value, and optimal use of land. Lagos particularly is one of the fastest growing cities in the world with Gross Domestic Product (GDP) and Internally Generated Revenue (IGR) in Nigeria, which suggests the need for multiple complex buildings, and the need for their maintenance cannot be overemphasised. This maintenance aspect requires tremendous work due to the complexity attached and several strategies springing up. Different studies reveal that both performance measurements and factors are essential aspects in evaluating maintenance management. Thus, this study seeks to explore performance elements that could improve maintenance. Personnel attitude, maintenance policy, maintenance review, and maintenance implementation were measured relative to computerised maintenance management system (CMMS) performance. With a random sampling technique, a sample of 134 Facility Management (FM) practitioners involved in high-rise office buildings was used to assess the effects of CMMS deployment. Results were analysed by Partial Least Squares Structural Equation Modelling (PLS-SEM). Findings from this study highlighted an indirect effect size and a large predictive relevance of personnel attitude as a critical factor for a smooth maintenance execution procedure of 12.59% and a standard operating procedure (SOP) of 15.64% on maintenance implementation to contribute 28.36% to performance. This paper uncovers the place of personnel attitude in determining effective maintenance.