TY - GEN
T1 - Feasibility of an Integrated Heuristic and Machine Learning Approach for Schedule Health Monitoring in Construction
AU - Jung, Yoonhwa
AU - Amer, Fouad
AU - Golparvar-Fard, Mani
N1 - Publisher Copyright:
© 2022 ASCE.
PY - 2022
Y1 - 2022
N2 - Project planning and controls requires planners to continuously revise project schedules to meet evolving requirements and constraints during a construction. Such an activity is usually performed under strict deadlines and planners are often forced to set aside good planning principles to deliver the updated schedules on time. To assist planners with validating their schedules, this paper explores the feasibility of using an integrated approach based on heuristics and machine learning methods to check the quality of a construction schedule. Specifically, building on the predefined rules and heuristics formulated in the Defense Contract Management Agency (DCMA)'s 14 Point Schedule Quality Assessment, this paper explores the feasibility of heuristic-based and deep learning methods to assess a project schedule health from qualitative and quantitative perspectives. Experimental results from thirty-five real-world projects are presented which demonstrate the feasibility of these underlying methods in highlighting schedule deviations from industry guidelines as well as following the best planning practices. A path forward toward a completely automated schedule health assessment system is discussed in detail.
AB - Project planning and controls requires planners to continuously revise project schedules to meet evolving requirements and constraints during a construction. Such an activity is usually performed under strict deadlines and planners are often forced to set aside good planning principles to deliver the updated schedules on time. To assist planners with validating their schedules, this paper explores the feasibility of using an integrated approach based on heuristics and machine learning methods to check the quality of a construction schedule. Specifically, building on the predefined rules and heuristics formulated in the Defense Contract Management Agency (DCMA)'s 14 Point Schedule Quality Assessment, this paper explores the feasibility of heuristic-based and deep learning methods to assess a project schedule health from qualitative and quantitative perspectives. Experimental results from thirty-five real-world projects are presented which demonstrate the feasibility of these underlying methods in highlighting schedule deviations from industry guidelines as well as following the best planning practices. A path forward toward a completely automated schedule health assessment system is discussed in detail.
UR - http://www.scopus.com/inward/record.url?scp=85128945273&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85128945273&partnerID=8YFLogxK
U2 - 10.1061/9780784483978.037
DO - 10.1061/9780784483978.037
M3 - Conference contribution
AN - SCOPUS:85128945273
T3 - Construction Research Congress 2022: Project Management and Delivery, Controls, and Design and Materials - Selected Papers from Construction Research Congress 2022
SP - 351
EP - 360
BT - Construction Research Congress 2022
A2 - Jazizadeh, Farrokh
A2 - Shealy, Tripp
A2 - Garvin, Michael J.
PB - American Society of Civil Engineers
T2 - Construction Research Congress 2022: Project Management and Delivery, Controls, and Design and Materials, CRC 2022
Y2 - 9 March 2022 through 12 March 2022
ER -