Offshore and onshore oil and gas facilities, petroleum facilities, and thermal power plants must run safely and without interruption. When equipment fails, production can stop without warning, output falls, maintenance costs rise, and safety and environmental risks increase. Companies therefore need a quantitative way to answer the following questions:
- Is each equipment item or process unit reliable enough, and how will its reliability change over the project life?
- How available is each equipment item or process unit when it is needed, in each year of the project life?
- If a failure occurs, how long will it take to restore operation?
- Which equipment items or process units have the largest effect on the reliability, availability and maintainability of the whole system, and how does that ranking change from year to year over the project life?
- What is RAM?
RAM stands for Reliability, Availability and Maintainability. ISO 20815:2018, the international standard for production assurance and reliability management in the petroleum, petrochemical and natural gas industries, defines the three terms as follows:
- Reliability: the ability to perform as required, without failure, for a given time interval and under given conditions.
- Availability: the ability to be in a state to perform as required. Availability can be stated for a point in time or as an average over a period.
- Maintainability: the ability to be retained in, or restored to, a state in which it can perform as required, under given conditions of use and maintenance.
- Common RAM terms
The table below explains the terms used in the rest of this article.
| Term | Explanation |
| MTBF (mean time between failures) | The average operating time between one failure and the next. A higher MTBF generally indicates better reliability. |
| MTTR (mean time to repair) | The average time needed to repair equipment and return it to service. A lower MTTR means faster recovery. |
| Downtime | The period during which equipment or a system cannot perform its required function due to failure, maintenance, waiting for spare parts, or other operational reasons. |
| 1oo2 (one out of two) | Two units or channels are installed, and one working unit is enough to maintain the function. Example: with two pumps, one running pump meets the required flow rate. |
| 2oo3 (two out of three) | Three units or channels are installed, and at least two must work or agree. Example: with three gas detectors, two must detect gas before a shutdown is triggered. |
| Duty/standby | One unit runs as the duty unit while the other stays on standby, ready to take over if the duty unit fails or is taken out for maintenance. |
- What a RAM study is used for
A RAM study is a quantitative analysis built on equipment failure and repair data. With a RAM model, companies can:
- Assess the reliability, availability and maintainability of individual equipment items, support systems, process sections, and the facility as a whole.
- Identify the equipment or systems responsible for the greatest production loss, downtime, or operational risk.
- Assess how redundancy configurations such as 1oo2, 2oo3, or duty/standby affect availability.
- Optimize maintenance strategy, spare parts, resources, and operating costs throughout the asset’s life.
- Calculate RAM figures over time for each machine, equipment item, or equipment group, work out how much each one contributes to plant-wide performance, and rank them so improvement effort goes where it has the most effect.
- Provide the RAM figures required by commercial contracts, insurance documents, or design documents that specify plant availability or capacity.
- Managing reliability and availability over the asset life cycle
RAM work continues for the whole life of the asset, well after the design-stage calculation. During operation, the company keeps the RAM model current alongside a reliability-centered maintenance (RCM) program and a living database. The database is updated every year, and also when new equipment is added or when the design, P&IDs, operating regime, or maintenance strategy change.
At the design stage, engineers typically build the RAM model from reliability handbooks and databases such as OREDA, equipment documentation, equipment lists, P&IDs, the operating and maintenance philosophies, and engineering assumptions. Once the facility is operating, they should calibrate the model with actual data from the computerized maintenance management system (CMMS), including MTBF, MTTR, maintenance schedules, failure records, maintenance manpower, spare parts, logistics, costs, and downtime.
The results of managing RAM throughout the life cycle can be used to:
- Track reliability and availability, and rank the criticality of each equipment item, subsystem, main system, and the whole facility, for each year of operation.
- Forecast production profiles, production performance, and production losses caused by downtime or availability limitations.
- Estimate annual operating and maintenance costs, including labor, spare parts, consumables, support services, transportation, mobilization/demobilization, and total maintenance cost by equipment item.
- Identify bad actors, meaning equipment or systems that often cause failures, downtime, or production losses, so the company can prioritize improvements to design, maintenance strategy, spare parts, or redundancy.
- When to carry out a RAM study
A RAM study can be carried out at several points in the life of a project or asset. The objectives, input data, and level of detail change with each stage:
| Stage | Purpose of the RAM study |
| Concept / pre-feasibility / feasibility stage | Compare technology options, system configurations, redundancy levels, and availability targets before finalizing the investment decision. |
| FEED / basic design stage | Assess the initial design, identify bottlenecks, and check whether 1oo2, 2oo3, and duty/standby configurations and the maintenance requirements suit the design. |
| Detailed design / EPC stage | Update the model with vendor data, P&ID, equipment lists, and operating/maintenance philosophy; optimize spare parts, maintenance accessibility, and redundancy configuration. |
| Commissioning / pre-operation stage | Synchronize as-built data, build the asset hierarchy, set up initial CMMS data, and verify system readiness before commercial operation. |
| Operation stage | Update the model with actual CMMS data, analyze downtime, bad actors, MTBF, and MTTR, and optimize maintenance, spare parts, and manpower strategy. |
| When changes or modifications occur | Assess how changes handled under management of change (MOC), such as new equipment, P&ID changes, capacity changes, or changes to operating philosophy or maintenance strategy, affect reliability and availability. |
| When repeated failures or high downtime occur | Identify the root causes reducing availability, evaluate the effectiveness of improvement measures, and support decisions on equipment replacement or upgrade. |
| Life-extension stage | Assess the impact of equipment aging, the risk of declining reliability, the need for overhaul or replacement, and the cost of sustaining operations in subsequent years. |
- Relevant standards and legal basis
ISO 20815:2018, Production assurance and reliability management, sets out how to plan and manage production assurance and reliability over the life of petroleum, petrochemical and natural gas facilities.
ISO 14224:2016, Collection and exchange of reliability and maintenance data for equipment, gives a standard format for equipment, failure, and maintenance data, which are the main inputs to a RAM model.
Vietnamese regulations for oil and gas and thermal power do not yet require a standalone RAM study. Several regulations do, however, require work that overlaps with RAM, such as safety management, risk assessment, operation, maintenance, inspection, and equipment integrity:
- Decree 45/2023/ND-CP requires petroleum contractors to prepare safety management documents, namely a safety management program, a risk assessment report, and an emergency response plan. It also requires the safety management system to include procedures for operation, incident response, and maintenance of facilities, machinery, and equipment.
- Circular 40/2018/TT-BCT, with its later amendments, sets out how safety management documents for petroleum activities are prepared and what they must contain.
- Decree 13/2011/ND-CP on the safety of onshore petroleum facilities, amended by Decree 25/2019/ND-CP, sets responsibilities for the safety, operation, maintenance, and repair of onshore petroleum facilities and for their safety management systems.
- Circular 02/2025/TT-BCT requires power generation facilities, including thermal power plants, to carry out a technical safety assessment covering maintenance and repair, analysis of incident causes, and operational risk. For electrical equipment on the national grid, owners that use condition-based or reliability-centered maintenance may set inspection intervals from condition analysis, up to a maximum of 72 months.
- Circular 05/2025/TT-BCT on transmission, distribution, and metering systems requires distribution companies to publish power supply reliability data on their websites.
In the railway sector, Decree 67/2026/ND-CP on FEED design for national and local railway projects lists availability, reliability, maintainability, and safety (RAMS) requirements among the contents of the FEED design.
- SEN’s RAM consulting services
SEN has worked in health, safety, and environment (HSE), process safety, and technical risk management since 2006, and can support clients with RAM studies for critical industrial assets.
The typical scope covers offshore oil and gas production and processing facilities, including FSOs and FPSOs; onshore oil and gas facilities; gas processing stations; petroleum depots and terminals; and thermal power plants.
Systems commonly studied include pumps, compressors, turbines, boilers, electrical systems, control systems, the emergency shutdown (ESD) system, the fire and gas (F&G) system, firewater systems, and other critical support systems.
A RAM study by SEN follows these steps:
- Review equipment and operating data.
- Build the asset hierarchy.
- Standardize failure and maintenance data.
- Build the RAM model.
- Calculate reliability, availability and maintainability.
- Rank equipment by criticality.
- Identify production bottlenecks.
- Recommend improvements to design, redundancy, maintenance, spare parts, logistics, cost, and resources.
8. Where to start
On a new project, start the RAM model at concept or FEED, while redundancy and equipment choices can still change. On an operating facility, start with the failure and downtime records in the CMMS, which show where availability is being lost. In both cases, update the model as the facility and its data change. To discuss a RAM study for your facility, contact SEN through www.senwork.com.