A return on investment analysis of applying health monitoring to LED lighting systems

LED lighting systems have become desirable because of their environmental and energy-saving advantages. The lack of information regarding LED reliability is a barrier to the further expansion of LED use, especially in large-scale applications such as street lighting and traffic lights, and safety-related applications such as automotive headlights. Prognostics and health management (PHM) techniques can be utilized to provide LED reliability information to remove this barrier. However, the return on investment (ROI) for LED lighting systems has been of concern. To reduce life cycle cost, a PHM maintenance approach with system health monitoring (SHM) is considered as a means of providing early warning of failure, reducing unscheduled maintenance events, and extending the time interval of maintenance cycles. This paper presents the ROI from a PHM maintenance approach with SHM in LED lighting systems compared with the unscheduled maintenance approach based on different exponential and normal failure distributions. Three different exponential distributions with 10%, 20%, and 30% failure rates were used to investigate how ROI changes with different failure rates. For each failure rate, the mean times to failure (MTTFs) were 41,000 h, 20,500 h, and 13,667 h, respectively. Three normal failure distributions with the same MTTFs as those of the exponential distributions were utilized to compare the results with the exponential distributions. ROI results showed that the PHM maintenance approach with SHM is required for cost savings in the exponential failure distributions. In case of the normal distributions, the PHM maintenance approach with SHM shows ROI benefits when MTTFs are less than 30,000 h. The PHM maintenance approach with SHM needs to be considered in industrial applications based on the reliability of LED lighting systems to maximize the ROI benefit when the total life cycle cost of the system employing the unscheduled maintenance is greater than the total life cycle cost of the system employing the PHM maintenance approach with SHM. (C) 2015 Elsevier Ltd. All rights reserved.