Structuring Competition for Networks of Buildings

This article appeared in Buildings Magazine on Octoboer 1, 2018.

Governments with networks of public buildings have unique opportunities to structure competition around building systems to produce both dramatic savings in life cycle cost and improvements in level of service. Examples of this include a school system, with multiple elementary, middle and high schools, is an example of a network of public buildings – operated under a unified revenue and expense model, and with common management – and a county court system.

Our past research at MIT indicates that 65 percent of what governments spend on infrastructure assets is optimal. That percentage moves dramatically up if:

  • Asset management systems are upgraded to include source and use benchmarking, and
  • Procurement systems are opened up to allow life cycle competition

The chance to save money improve service arises because networks contains multiple instances of the same systems in each building. Systems common to public buildings include:

  • Roofing
  • Air conditioning
  • Heating
  • Pavement
  • Hot water
  • Elevators
  • Lighting
  • Exterior window/wall
  • Fire safety
  • Security (including access control)
  • Telecom (including internet)
  • Signage


Optimize a System Across the Network: An Example

Unlike public owners operating a single facility, facility managers running a network of public buildings have the chance to plan how to optimize one system at a time, across the network.

This involves moving from a single procurement strategy that focuses on one building at a time to a broader approach that includes a second strategy focused on improving service results across the network in each of the systems listed above.

With roofing, this second strategy provides value for money (higher level of service at a lower life cycle cost). At any given time, each school roof in a school system is in a different state of repair. A few might not need any maintenance or repair, while others might require minor repairs, major repairs or substantial replacements. The condition of individual roofs in the network changes over time – but the distribution of conditions across the network is more stable. Regular maintenance is required every year – on different roofs – along with a predictable mix of minor repairs, major repairs and replacements.

The nature of network roofing needs presents opportunities for facility managers to package roofing projects differently over the long term so that “roofing” consumes a predictable portion of resources each year from the total pool of available resources.

Network analysis gives managers the chance to answer different, more helpful questions: For each building, how have different roofing systems performed – in terms of:

  • Initial cost
  • Ongoing repair cost
  • Life cycle cost
  • Longevity

Make Informed Decisions With Data

Armed with accurate long-term data about the performance of different systems on each building in the network, facility managers are in a position to build predictive maintenance, repair, and replacement models for roofing expenses across the network.

Predictive models rely on basic financial analysis of source and use of funds. They permit public owners to benchmark one specific or type of roof against another. Importantly, comparative source and use analysis allows government to use competitive procurement to independently verify that money for roofs is spent in the right place, at the right time, and in the right way to meet the level of service required across the network.

Each system provides similar opportunities to work on the central problem in asset management:  how to allocate scarce resources across an infrastructure network when those resources have alternate uses.

An accurate, historical record of source and use of infrastructure funds provides a solid foundation for robust predictive models. These models, in turn, allow procurements to be structured to keep competitive pressure on private sector providers to accomplish the only two things that matter:

  • Higher levels of service
  • Lower lifecycle cost

Simple spreadsheet tools[1] are used to build these predictive models and scenarios using them.

Think Long Term

In the real world, procurement strategy for a school or government system is dependent on the condition of the assets and the historical record of sources and uses of funds. For relatively stable networks – like these – opening procurement models to allow 10-15-year contracts to maintain, repair, and replace systems is a proven approach.

Structured competitions for network of building systems dramatically improves how public funds are spent on infrastructure.


By:  Dr. John Brown Miller

Miller was a reporter on the ABA’s 2007 Model Code for Public Infrastructure Procurement project, providing best practices in procurement to state and local governments. You can find him on LinkedIn and Twitter @JohnBrownMiller.

[1] See, for example, ENTERPRISE RISK MANAGEMENT IN INFRASTRUCTURE – PART 3, August 1, 2018, civil + structural Engineer.