{{Article.Title}}

{{Article.SubTitle}}

By {{Article.AuthorName}} | {{Article.PublicationDate.slice(6, -2) | date:'EEEE, MMMM d, y'}}
{{TotalFavorites}} Favorite{{TotalFavorites>1? 's' : ''}}
{{Article.Caption}}

Capital projects are both complex and resource intensive. When the risk and likelihood of overrun is high, how a company chooses to manage its project portfolio can have a substantial impact on the bottom line. Missing an opportunity to save millions of dollars’ worth of cost may mean the difference between black and red on the annual balance sheet. A mathematical approach to portfolio planning, using available AI-based tools, helps to avoid missed opportunities by accounting for factors that other methods just can’t consider.

At enterprise-level organizations, there can be hundreds of capital projects in a portfolio at any given time. Some of these are related to the growth plans of the company—such as opening new locations or production facilities—while others may be about repairing existing assets or various building improvements.

Most often, the executives who make up the planning team decide the order in which projects are undertaken. Although they may use a few spreadsheets here and there, decisions are primarily based on the instincts and experience of these team members. 

It’s important, of course, not to disregard the value of intuition and seniority. Still, how do they know these prioritized projects will really add as much shareholder value as something else would have? More stakeholders are realizing that when a capital project portfolio reaches a certain level—whether through sheer number of projects or the size of those projects or both—having an objective tool in place that can support the leadership team’s decision-making process is a critical resource. 

Optimizing decision making in the built environment

A mathematical approach to portfolio planning considers the relationships between projects. It pinpoints where dramatic reductions in cost and risk can be achieved through smart sequencing of projects. It maximizes business value across the organization’s investments and accounts for all the variables in play—time, resources and financials. 

Taking a more quantitative look at project planning and sequencing can be done in a number of different ways. Some are purely financial (which, among the companies using any kind of decision support at all, tends to be the most commonly used method in construction today). This approach ignores time and resource-related factors, though, so it doesn’t tend to generate maximum return on investment. 

Another method is weighted factor scoring using a decision committee. Each member considers an array of different factors, then estimates the importance of each according to his or her own judgment and experience. This is a more holistic approach than considering financial factors alone, but it’s still inherently subjective. 

The final (and arguably most desirable) method is one that utilizes the optimization model—typically with some form of mathematical programming. An optimization model takes into account a wide array of variables to determine the set and/or sequence that will deliver the maximum benefit. Artificial intelligence tools would fall under this category. AI can determine which of hundreds of projects will add the most value based on a wide array of and in which sequence they should be done. 

Paving the way for AI

There are already a number of AI-based tools that can provide decision-making support. In order to use them, however, most AEC organizations must first take the interim step of becoming more data-centric, data savvy and willing to share information across departmental silos. 

One of the biggest benefits of an objective approach to portfolio planning is, in fact, that it gets valuable information out of documents and makes it more useful to the organization as a whole. 

Too often data is not shared—even within the same organization—for arbitrary or territorial reasons. Shifting to single-source-of-truth style systems and models not only fosters greater transparency, it’s what’s driving the most value for organizations today. Sharing data also makes a mathematical approach to portfolio optimization possible. On the facilities management side of things, having a more data-centric approach means organizations can start to tie into external data systems of weather and financial information, which can help flag potentially anomalous events, prevent loss of life and major unforeseen expenses. 

Imagine for instance, if AI had been able to tell the management team at Bellevue Hospital that there was an 85% chance the basement would flood before Hurricane Sandy hit New York City in 2012. The hospital might have been able to evacuate its 736 patients when the power was still on and the circumstances much safer.

Looking ahead

A numerically-driven decision support system can help teams determine the right path forward when a great amount of complexity and risk is involved. Organizations that embrace a data-centric model stand to gain a competitive edge in the form of smarter, more value-added portfolio sequencing. Leveraging AI, they’ll also set themselves up to significantly reduce overhead and boost the organization’s decision-making capabilities as a whole.

Print

 Comments ({{Comments.length}})

  • {{comment.Name}}

    {{comment.Text}}

    {{comment.DateCreated.slice(6, -2) | date: 'MMM d, y h:mm:ss a'}}

Leave a comment

Required!
Required! Not valid email!
Required!