Delivering Reliability Centered Maintenance in Railroads

Ever since Nolan and Heap completed their report into United Airlines in 1978, asset intensive organizations have sought to apply the principles of Reliability Centered Maintenance (RCM), pursuing the goal of better safety and reliability and reduced costs. Network Rail began this journey in 2005 and whilst it has been ultimately successful in improving operational performance, it took longer than hoped. This article will explore some of the factors behind that.

The principles of RCM are well documented in John Moubray’s seminal 1999 book entitled RCMII Relaibility-centered Maintenance and subsequently developed into the concept of RCM3 by Marius Basson in his 2018 book entitled RCM3 Risk-based Reliability Centered Maintenance. However, there is more to undertaking a fundamental rewrite of your maintenance procedures than the technical application of the RCM process.

In 2005, Network Rail assigned a 2-person project team to deliver an initial tranche of RCM analyses for signalling equipment, with the bulk of the technical resources being borrowed from elsewhere in the business. Progress was slow due to the lack of availability of the subject matter experts and in 2008, the decision was made to assign a team of technical experts to the task full-time and create a focused and full-time project team to deliver RoSE (Reliability of Signalling Equipment).

By 2011, a range of signalling asset types had new maintenance regimes available, and these had been applied to 125,000 individual assets with another 30,000 awaiting implementation. This latter figure reveals one of the time-consuming elements of deploying RCM: the pre-conditions, which we will revisit later in the article.

One of the other critical dependencies for progressing the program was the approval of the owners of the maintenance standards. Whilst they believed in the benefits of the program, and in the RCM process, they were ultimately cautious about signing off the new task lists.

The RCM process doesn’t really address pre-conditions, but we observed that across assets of the same type, there will be a spectrum states of repair. Assets in a good state of repair can transition to the new RCM regime immediately, whereas those in poorer condition need to remain on their original regime until they can be brought up to a sufficiently good state to transition to the new regime.

Whilst the solution may appear to be a purely technical task to perform a condition assessment, we found that the motivation to complete, or not to complete, this assessment was strongly linked to the supervisor’s and manager’s support for the RCM program. The same issue afflicted the assessments themselves with some maintenance technicians deliberately failing to complete their assessments for fear that they were a threat to their future work.

Successfully overcoming the non-technical challenges to the program resulted in around 367,000 signalling assets transitioning to RCM, saving 438,000 maintenance hours per year whilst observing an overall increase in reliability.

We started with signalling assets because the RCM process is directly applicable; asset types are consistent and their operating context does not impact the P-F interval (Potential Failure to Functional Failure) and hence the maintenance frequency. For track assets, we identified that at a track system level reliability is driven by different combinations of similar asset types (ties, fastenings, rail, etc.) in differing operating contexts (for example linespeed and tonnage). To overcome this challenge, we produced a set of template analyses that informed the RCM process undertaken for a geographic area with first-principles analysis being applied to anything not covered by the template.

By combining risk-based RCM techniques with technology led high quality track condition and performance data, over 95% of Network Rail’s track assets are now managed by RCM. This approach delivers significant condition and performance benefits, efficiency gains and safety improvements both from reducing rail defects but also by removing the requirement for manual track inspection in live traffic.

Overhead Catenary Systems (OCS) present yet another set of variables, with the asset types being broadly consistent with widely varying operating contexts, such as speed and service frequency. In this case, we developed regimes with standardised tasks, mandated at differing frequencies dictated by the operating context and calculated through a weighted matrix.

At Network Rail Consulting, we have the experience of delivering a successful Reliability Centered Maintenance Program for an operating railroad and apply this proven capability to help other agencies optimize their networks. Our services include developing a maintenance optimization strategy, delivering a program of RCM analyses, undertaking change management to embed the technical changes and measuring the benefits of the program against the original business case.

Ready to optimise your maintenance strategy? Contact us today to implement proven RCM solutions that deliver efficiency, reliability, and measurable results.