Capital Project Management: Where Reliability Gets Bought, Built In, or Lost

A capital project is where a plant's next twenty years of reliability are decided. Specifications are written, vendors are chosen, equipment is installed, and the asset is handed over to the people who will live with it. Each of those steps either protects or erodes inherent reliability.

This is the sixth article in our Reliability Engineering for Maintenance series. It spans Stages 1 to 3 of the full P-F curve: design, receiving, and installation. It builds on reliability centered design by showing how those intentions survive, or don't, the trip from drawing to operation.

Why projects lose reliability

Capital projects are measured on cost, schedule, and scope. Reliability after handover is rarely on the scorecard. The pressure to finish on time leads to predictable compromises: cheaper equipment than specified, skipped acceptance tests, rushed commissioning, and incomplete documentation. The project team moves on, and the operating team inherits the consequences.

Where reliability can be protected

Front-end definition. Early scoping is where reliability and maintainability requirements are cheapest to include. Define the operating duty, availability expectations, and monitoring needs before the design is frozen.

Specifications and vendor selection. Specify tolerances, balance and alignment requirements, materials, lubrication, and monitoring provisions. Evaluate bids on total cost of ownership, including spares, energy, and expected maintenance, not just purchase price.

Fabrication and acceptance testing. Witness factory acceptance tests for critical equipment. Defects found at the vendor are far cheaper to correct than defects found in service.

Receiving and storage. Inspect on arrival, store correctly, and control substitutions, so the equipment installed matches what was specified.

Installation and commissioning. Verify alignment, foundation and grout condition, pipe strain, torque, cleanliness, and lubrication against defined acceptance criteria. Record baseline vibration, temperature, and electrical signatures. These baselines are the reference point for every condition-monitoring reading that follows.

Handover. A clean handover gives operations a usable asset, not just a running one.

What a good handover includes

• Assets loaded into the asset register and hierarchy with consistent naming
• Criticality rankings for the new equipment
• Maintenance strategies and PM tasks developed from the failure modes, not copied from the vendor manual by default
• Spare parts and BOMs identified, with critical spares on hand
• Manuals, drawings, and as-built documentation
• Baseline condition data
• Trained operators and technicians
• A clear plan for using the warranty period to find and resolve early defects

Common mistakes

• Reliability and maintenance people brought in only at the end, when changes are costly
• Skipping or shortening commissioning checks to save schedule
• No baseline data, which removes your reference for future condition monitoring
• Handover without spares or strategies, so the plant starts out reactive
• No post-project review comparing what was promised with how the asset performs

Treat handover as a gate

The most effective change is organizational: make reliability deliverables a condition of project completion. If the asset register, strategies, spares, baselines, and training aren't ready, the project isn't finished, however good it looks on schedule.