Assessment of integrity of reasoning in large-scale decision systems: application to public transit investment project evaluation

Author(s)Mangalpally, Sharat C.
Date Accessioned2020-08-05T17:39:31Z
Date Available2020-08-05T17:39:31Z
Publication Date2005
AbstractThis thesis presents an approach to measure the integrity of reasoning process of large-scale public projects, such as civil infrastructure projects, and public transportation projects. Decision-making in public projects requires careful examination of the magnitude of impacts, cost responsibilities and benefits enjoyed by various stakeholders. The analysis of public projects is characterized by uncertainty associated with the large number of interactions among the system variables, subjective nature of the goals and objectives that are often interrelated, and the presence of many consequences and concerns of different population groups. The proposed method captures the uncertainty in the reasoning process during the project planning. ☐ A four-step analysis is proposed to deal with uncertainty. First, the system’s objectives are defined and the variables related to each objective are identified. Second, the system is decomposed into smaller subsystems based on each objective. These subsystems are developed such that the variables are all connected in a logical sequence of cause-effect relationships that originate from the initial conditions and lead to the objectives of the overall project. Third, the logical validity of the subsystems is examined by a group of experts from different backgrounds by providing the strength of agreement for each causal relationship expressed as the confidence measure. Finally the confidence measures of the subsystems are aggregated and the overall confidence in the performance of an alternative in achieving the objectives is obtained. ☐ The study applies the proposed method to evaluate a monorail project in the northern New Castle County of Delaware. The variables involved in the monorail project were identified and the causal network was constructed. Each causal relationship was evaluated using confidence measures obtained by interviewing a group of experts. The overall strength of reasoning for achieving an objective is calculated through aggregation of confidence measures. Considering the interdependency of the objectives, the confidence measure of achieving set of all the objectives by a given alternative is computed. ☐ The potential benefit of the proposed method is in the process of evaluating the transit alternatives as prescribed by the FTA (Federal Transit Administration) for the New Starts Projects. The method does not point to a specific alternative as a solution, instead, it provides a set of numbers that reflect the strength of reasoning about the achievement of the desired outcome by each alternative. This will help the decision makers understand the opinions of different cross sections of the people and assess the need for additional studies and restructuring the reasoning network. The proposed method hence enforces more integrity and objective evaluation in the project funding decision.en_US
AdvisorKikuchi, Shinya
DegreeM.C.E.
DepartmentUniversity of Delaware, Department of Civil and Environmental Engineering
Unique Identifier69869016
URLhttps://udspace.udel.edu/handle/19716/27364
PublisherUniversity of Delawareen_US
URIhttps://search.proquest.com/docview/304993204?accountid=10457
dc.subject.lcshProject management -- Decision making
dc.subject.lcshTransportation -- Planning
dc.subject.lcshUncertainty (Information theory)
dc.subject.lcshMonorail railroads -- Delaware -- New Castle County -- Planning
TitleAssessment of integrity of reasoning in large-scale decision systems: application to public transit investment project evaluationen_US
TypeThesisen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Assessment_of_integrity_of_rea.pdf
Size:
1.07 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.22 KB
Format:
Item-specific license agreed upon to submission
Description: