Too often site maintenance programs are implemented independently of the facility construction effort and, in many cases, almost as an afterthought.

There are three fundamental problems with this strategy.

  • The maintenance program needs to be fully functioning on day one when the site goes live.
  • Many decisions and strategies made regarding the maintenance program can directly influence how the physical facility is best designed. 
  • Developing and implementing the site maintenance program concurrently and in conjunction with the facility construction project can lead to significant opportunities for synergy and efficiency.

This is especially true in the deployment and implementation of what has become the de facto standard foundation of the maintenance process; the computerized maintenance management system, or CMMS.

The best approach is to incorporate the design, development, and implementation of the maintenance program in the design, construction, and commissioning of the physical facility. This includes establishing the objectives, requirements, strategies, and performance criteria of the maintenance program in the same site programming effort and owner’s project requirements (OPR) document as used to capture and define the same aspects of the physical facility. Similarly, the development of the basis-of-design (BOD) document should incorporate aspects of what “means and methods” will be employed by the maintenance program to meet the OPR maintenance objectives.

The construction documents (both drawings and specifications) should define contractor efforts, vendor products, and deliverables that are essentially maintenance offshoots of what is being installed and constructed. And just as startup and testing of the physical infrastructure of systems and equipment is performed as part of the commissioning acceptance phase, the same can be done for CMMS and associated maintenance programs. And a final match—a sound maintenance program and CMMS requires site-specific policies and procedures, training, and documentation as is required for the actual infrastructure to be maintained. So let’s break this process down into more detail and see where those opportunities for synergies and efficiencies lay.

Start with the programming phase. This includes making many high-level decisions that will provide direction and strategies that are critical to developing a sound maintenance program. Some examples of maintenance-related considerations include the following questions: What maintenance will be performed in-house and what will be out-sourced? Will the CMMS database be integrated with other similar databases such as asset management (inventory control) systems, configuration management systems, workflow (move-add-change) systems, etc.? Will the maintenance program incorporate reliability-centered maintenance processes? Will assets be bar-coded and hand-held scanners, electronic tablets, and PDAs be employed? Where will the maintenance department reside within the site/corporate management structure, and how will it be organized?

Will the maintenance department be combined or separate from the operations department? Will the site have maintenance staff on-site 24/7? Will there be separate critical and non-critical programs with distinctly different protocols? Will the facilities O&M staff report up through IT or not, and will service-level agreements be used between facilities and end-users, etc.? The decisions made regarding these questions will not only affect how the maintenance program gets designed and resourced but also have direct and indirect implications on how the physical facility gets designed and ultimately managed.

Now let’s proceed to the design phase of the maintenance program. As with the facility engineering and design, start with establishing the “means and methods” in the BOD. This is where the CMMS requirements and performance criteria get defined, including hardware (server, electronic storage/archive, operator workstation, remote access, etc.). this is not unlike how the facility building automation system and other computer-based monitoring and controls get defined. The BOD can specify a preferred CMMS provider and products and whether “substitutions of equal performance and capabilities can be submitted for approval,” as is done for other infrastructure.

The BOD should also establish physical aspects of the maintenance program such as where the maintenance staff resides; what maintenance spaces are required and where they are situated; and what maintenance shops, tools, equipment, and power are required such as paint shop, carpenter shop, welding/machine shop, electrical and electronics test shop, and rigging and transport equipment (jib cranes, forklift, dollies, come-alongs, etc.). Some of these spaces should include emergency power, since they need to function during emergencies and utility outages, and also may require substantial power outlets of varying voltages. Other maintenance requirements might include landscaping, hazmat and flammable storage facilities, stock-rooms, and environmentally controlled storage for sensitive gear and test equipment.

It becomes obvious why maintenance program decisions made during the programming phase impact the physical facility design. If all or most maintenance is out-sourced, then many of the spaces and requirements listed above may not be applicable. Outside contractors and vendors can be required to have and maintain test equipment, inventory of parts, rigging equipment, etc. With in-house staff (direct employees and/or permanent, in-house contractors), the site must accommodate these requirements. Failure to think this through before the design is substantially complete leaves the hard choice of adopting the maintenance program to the facility design or modifying the design to accommodate the maintenance program. Either way, otherwise unnecessary compromises must be made.

Once the maintenance program is embedded in the BOD, the construction documents (CDs) specify what the contractor has to deliver, and this is where the real opportunities for synergies are found. Studies have shown that many CMMS initiatives have either failed outright or significantly come short of meeting expectations due to underestimating the level of effort and costs required to setup, populate, implement, and provide training for the system.

Many of the pitfalls can be avoided when the CMMS installation and deployment is combined with, and incorporated into, the infrastructure design and CDs. The BOD and CDs can establish naming and labeling conventions that carry into the CMMS. Equipment purchase orders can include delivery of CMMS data collection sheets, in addition to the standard O&M manuals and documents that are typically included.

These would include model number, serial number, consumables, special tooling requirements, local vendor contact info, maintenance procedures, and recommended frequencies, etc. With a little innovation, some of the data could be delivered electronically and imported directly into the CMMS database. Equipment IDs and/or names, locations, areas served, and other details that would be input into a CMMS could be made to match (and therefore taken directly from) the CD single-line diagrams and equipment/panel schedules.

There would be no need for cross referencing or lookup tables. And the naming and ID conventions should be logical and structured so the ID alone would convey the building, floor, wing, room, unit type, unit number, system, and whether critical or non-critical service.

When done properly, this allows the CMMS to easily sort and filter assets for various report and tracking functions. This can increase the usefulness of the CMMS including producing a multi-year capital plan, combining and coordinating similar maintenance activities to occur concurrently or, in some cases, avoid occurring concurrently as the case may be, and to assign spare parts inventory to units that are of the same model and many other purposes.

The developed maintenance program should be tested and verified for functionality and accuracy, or in other words, it should undergo functional testing. The CMMS system needs to go through a thorough vendor startup and checkout. Sample, or even initial asset data, should be entered, maintenance tasks generated, results entered, and reports generated. In short, it needs to be put through its paces and debugged prior to proceeding with full population with all site input data. Maintenance staff should also be provided training in onsite policies and procedures, work authorization and reporting processes, and actual maintenance procedures for the installed equipment. This training should incorporate systems/equipment academics, practice and testing (demonstration) on performing maintenance tasks, and on performing the required operations actions required to place critical systems into maintenance mode and safety processes that make the equipment safe to work on.

Safety practices require a greater magnitude effort in complex critical infrastructure than typical commercial office buildings. Many systems and equipment are provided multiple power sources, stored energy, and automation that create increased risk for accidents. Safe practices must combine standard safety processes and program requirements with site-specific training and procedures so the actual hazards are first identified, second understood, and third mitigated or eliminated without “accidentally” causing unintended shutdown to redundant or backup systems and impact to critical operations. In other words, what’s needed is “discipline, rigor, experience, training, process driven procedures, and a culture of excellence.”

An interesting question is how much maintenance is the right amount of maintenance. Too little is obviously not good, but does “the more the better” hold true? Recent studies have shown that too much maintenance can actually degrade equipment in some circumstances, increase operational risk associated with system reconfiguration above the risk due to maintenance preventable failure, and waste both manpower and money for little or no benefit. So what’s the answer? The answer is reliability-centered maintenance (RCM).

RCM is a topic worthy of an article of its own, but in general it is a holistic approach to maintenance that considers the purpose and operational value of the asset, the impact of failure, viable failure mechanisms (cause and root-cause), means to assess the actual condition (on-line condition monitoring), predictive maintenance technologies processes, and an objective decision process to determine the appropriate maintenance taking all of the above into consideration. Some assets might be allowed to “run to failure” since they are expensive to maintain, easily replaced, and failure is of little consequence. Other assets may be considered necessary at all costs. Another aspect of RCM is it includes continuous process improvement where the results of the maintenance process are continuously assessed and evaluated and adjusted to improve results. The maintenance program itself needs to be maintained.