How can you measure and improve system reliability?

In order to understand and define reliability requirements it is important to take a holistic lifecycle view. All stages within an asset or products lifecycle impact reliability and efforts should be made to manage each lifecycle stage in a way that contributes towards reliability. It is important to note that as the lifecycle stages advance from initial concept, design, installation, testing, commissioning, operations, maintenance and ultimate disposal of the asset; the ability to positively influence reliability diminishes. This makes it imperative that reliability targets should be planned and followed through early lifecycle stages.

Consider taxonomy or hierarchy structure. 1. Process, 2. System, 3. Equipment, 4. Component and 5. Part See this from this perspective: Reliability of the component/parts; Availability of the Equipment. From 1-2, 'Process' and 'System' reliability can be improved by introducing redundancy. From 3-5, this is the OEM scope where reliability is inherent in design. 4-5 is where the use of derated components and parts come in as well as superior materials and stress-free parts and components are considered. For electric and electronics, redundancy circuitry at component level maybe considered to improve reliability. Similarly, if the equipment has sub-systems where redundant system might be put in place to ensure sub-system reliability.

Design/Process/Maintenance Reliability. Reliability = f (Stress, Process Parameters, Materials, Usage). Measure Reliability to Support & Manage Process Inputs during the Design, Operate and Maintenance Phase. RAM Modelling during the Design phase/EPC Phase/O&M Phase will help the Organization to understand better about the Performance of the assets and System losses.

1. Build risk profile of production asset - (risks gathered from equipment strategy, SHE & Financial exposure inventory, Vulnerability list, etc.) 2 Perform risk screening by identifying risks resided at the "Red Zone" 3. Conduct deep dive of each critical risk and determine the mitigating solution. 4. Perform Benefit to Cost Analysis for prioritization. 5. Select only those risk items within the cut rate and presented to management for approval

Measuring reliability KPI includes measuring the system losses by unscheduled incidents across the system. Example for an process plant the main product losses led to great production losses which affects the plant revenue and costs and this can be measured and recorded and reported to the management so that Management will picture of unavailability and reliability improvements on the system. Shell does this kind of losses tracking and measure the plant utilization percentage by monthly opening plan (MOP) VS actual production. Unplanned DOWNTIME % can be measured and benchmarking can be studied. Major UPDT incidents can be captured with respect to system, equipment and process and thus can be monitored and improved for an system/asset.

RCM livin programs for an asset improves and reviews the RCM for an system. Existing RCM strategies will still have some new failure modes, so by doing continues improvement activity for an asset on RCM livin programs makes to identify the system failures and finding a way to do RCA, FMEA, CAUSAL INVESTIGATION STUDIES, thus improves asset reliability standards.