Enhancing Decision-Making on Sustainable Building Projects Using Influence Diagrams

Project Team

Michael Lepech, Assistant Professor, Civil and Environmental Engineering

Ross Shachter, Associate Professor, Management Science and Engineering

Kelcie Abraham, Ph.D. Candidate, Sustainable Design & Construction


Architecture, engineering, and construction teams need more effective decision-making methods and visualization tools, particularly during early-stage design and pre-construction of sustainability-focused projects when decisions have the most impact on sustainability performance. The multi-disciplinary nature of these decisions and the engagement of multiple stakeholders often result in decision problems with multiple objectives and confusing alternatives, necessitating decision-making approaches that clearly represent the rich data environment owners and AEC professionals face. This work proposes an approach based on influence diagrams. While influence diagrams have been used in other industries (e.g. environmental management) with great success, they have seldom been applied to the built environment. This research will evaluate the ability of decision-making visualization methods to enhance sustainability objectives. Ultimately, by better integrating visualization methods into decision-making and developing more rigorous techniques that are quick and easy to use, AEC professionals can help clients achieve sustainability goals with greater efficiency and clarity.

Project Background

Research Motivation

Over the course of a building project, architecture, engineering, and construction (AEC) consultants predict and evaluate the performance of many different design and construction options. The daily design-construction recommendations that these AEC consultants make to the client have significant impacts on the sustainability of the building throughout its life cycle. (Ugwu and Haupt 2007) With contractors and designers frequently working under strict budget and schedule constraints, AEC consultants need superior planning, design, and construction processes to achieve client goals – and the tools to communicate these processes and the options they consider in a clear, effective fashion. It is also critical that AEC professionals have the appropriate tools and effective visualization tools to explain sustainability options and pathways to owners and clients that have limited knowledge and experience with sustainability issues and the ways by which sustainable built environments can be constructed.

One of the most active and important stages for decision-making and choice representation to owners and clients is during the design development and construction documents stages of the building lifecycle. This means that during design, decision makers are dealing with decisions that range from a relatively large number of options (which can be hard to explain to owners) to those that have fewer, better-defined alternatives to consider (but require a more detailed analysis and can be equally hard to explain to owners). Thus, the ability to model and visually represent the decision-making processes associated with sustainable construction is similarly useful across throughout the design development timeline even as the ability of decision makers to change project design and cost decreases and the cost of design changes (in terms of resources and budget) increases significantly. 

Influence Diagrams

One way to visually represent a decision-making process is through influence diagrams. Influence diagrams have been introduced and used in a number of complex decision-making scenarios (Shachter 1986). The theory of influence diagrams dates back to the early 1980s, and a variety of commercial software packages are on the market. An influence diagram is an expansion of a decision tree: a graphical structure with directed arcs and with decision, chance and expected value nodes. Within these diagrams, hierarchical structure is not required, only that directed cycles of decision-making are not permitted. The approach is both straightforward and practical in structuring and constructing models, in performing optimization, sensitivity tests, and value of information analysis, along with analyzing management problems under various prevailing risk models. A simple influence diagram for building life cycle performance is shown below. Within this diagram, the actual design performance of the building over its lifetime is uncertain, but design performance predictions may be observed before a final decision is made about the building design. Ultimately, the life cycle impact of the building is determined by the design decisions and the actual performance.


To date, influence diagrams have been applied to several problems in Civil and Environmental Engineering, including environmental management (Varis et al. 1990; Varis 1997; Kuikka et al. 1999), siting of hazardous waste management facilities (Merkhofer et al. 1997), and more recently energy plant siting (Zhou et al. 2006). Influence diagrams allow probabilistic, Bayesian studies with classical decision analytic concepts such as risk attitude analysis, value of information and control, multi-attribute analysis, and various structural analyses. Such diagrams can serve as a crucial link between higher-level decision-making visualization tools (i.e. Pareto fronts) and the operational decisions that owners and AEC professionals are faced with on a regular basis.

Research Objectives

This research has two primary goals. The first is to determine how influence diagram visualization tools can support decision-making methods that produce the highest quality decisions – that is, how well are they suited for complex decision-making on sustainability-focused projects. The second objective is to evaluate the potential for using influence diagram techniques to integrate and communicate sustainability objectives within design and construction decision-making and in turn, to establish a foundation for further research to determine how integrated sustainability decision-making can be enhanced through visualization of complex decision processes.

Original Research Proposal

Proposal2014-05.WebVersion.pdf705.56 KB

Last modified Wed, 8 Apr, 2015 at 8:23