Visualizing and interacting with construction project performance information

Project engineers/managers need individually relevant and customizable information in order to monitor and assess the performance of construction projects. Current project management tools do not meet this need completely; contributing to both lack of information in certain aspects and information overload in others. The results from a contextual inquiry; including a set of interviews with different project members highlight that 60% of the information needs of the project members are not met. In order to meet such needs, relevant data needs to be processed to the desired level of detail and be displayed in the right format that fits to the characteristic of the data and the task that the project member is trying to do (e.g., compare, analyze the trend, etc.). In addition, desired interactions with the data need to be supported. The advent of information visualization techniques could help in addressing such needs and creating new opportunities to overcome information overload by enabling effective understanding of the datasets. However, there is no comprehensive study that formalizes how various visualization techniques can be represented, reasoned and visualized to meet the unique needs of different project members in the construction industry. As a step towards improving the visualization of and interaction with individually relevant project performance information, I have developed a framework that provides effective visual representations to project members. Identifying and addressing the individual information visualization needs of construction project team members are challenging since these needs differ in terms of the performance metrics that are of interest, the level of detail and content of the available data, and the visual representations that support the task and suitable to the data being displayed. With the goal of addressing these challenges, the research described in this thesis has had three objectives: (1) identifying the commonalities and differences in information visualization requirements of project members when monitoring and assessing the performance of construction projects, (2) developing a framework for defining, generating and visualizing individually relevant project performance information, and (3) formalizing the selection of visualization of individually relevant information for effectively monitoring the performance of construction projects. The research approach involved contextual design method with semi-structured interviews and direct observations of construction project members working on different projects to identify the commonalities and differences in their information visualization needs. The research tasks include consolidation of interview results, synthesizing literature on information visualization, prototyping for testing purposes, representing and navigating project performance data, developing metadata for project data, and representing and reasoning about selection of appropriate and effective visual representations. The validation studies include cross-analysis of interview results, user tests and a synthetic test to evaluate (1) the commonalities and differences in information visualization needs of project members, (2) the usability of the components of the visualization framework and, (3) the precision, usability and usefulness of the formalism in terms of retrieving effective visual representations. The contributions of the research include; (1) identification of commonalities and differences in information visualization needs of various project members, (2) development of a generic construction project performance visualization framework, and (3) development of a formalism that can enable project members to select and visualize individually relevant project performance information in an effective way. In terms of practical implications, the proposed research can be seen as a step towards improving the current way of performance monitoring by better utilization of visualization techniques leading to possible reduction in time for visualizing re ated information, and contributing to effective decision-making.