{"id":243,"date":"2025-08-09T21:32:36","date_gmt":"2025-08-10T01:32:36","guid":{"rendered":"http:\/\/vibe.test\/?p=243"},"modified":"2025-08-23T15:37:41","modified_gmt":"2025-08-23T19:37:41","slug":"infrastructure-cost-escalation-and-overrun","status":"publish","type":"post","link":"https:\/\/research.gsd.harvard.edu\/vibelab\/2025\/08\/09\/infrastructure-cost-escalation-and-overrun\/","title":{"rendered":"Infrastructure cost escalation and overrun"},"content":{"rendered":"

Infrastructure cost escalation and overrun<\/h1>\n\n\n

This project builds open Dr. Voulgaris\u2019s 2017 doctoral dissertation on optimism bias in cost estimation and ridership forecasting for transit infrastructure. It is well-known that infrastructure projects are commonly completed at a higher cost than is initially anticipated in the early stages of project planning. Each individual project has its own unique story with regard to why and how projected costs increase over the course of project development and construction. The purpose of this project is to uncover patterns with regard to when<\/em> or at what points in the project developments and construction life cycle costs tend to increase, with a specific focus on federally-funded transit infrastructure in the United States. <\/p>\n\n\n\n

While prior studies have documented the large differences between initial cost estimates and final project costs, there is a lack of research on whether costs have tended to increase during the early stages of project planning, during the later design or land acquisition stages, or after project construction begins. The timing of cost increases have important implications: the earlier cost escalations arise, the better opportunity project sponsors may have to abandon projects that may no longer be cost-effective. While the histories of federally-funded transit projects in the United States are generally well- and publicly-documented in the form of published narratives, the lack of a comprehensive database of the timing of cost estimate increases has been a barrier to a systematic review of when cost increases commonly arise. This project began as an effort to assemble such a database manually. With the emergence of large-language models, we are developing a method to use these models to generate structured datasets on transit cost escalation from existing narrative data sources.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n<\/div>\n\n\n\n
\n

ViBE Lab Principal Investigator<\/h2>\n\n\n\n