An activity and flow-based construction model for managing on-site work

Placeholder Show Content

Abstract/Contents

Abstract
Construction field managers often struggle to keep projects on schedule, resulting in time and cost overruns. Schedule conformance depends on the activities starting and finishing on time. However, activities are often delayed because the flows necessary to start their execution are unavailable. These flows can be classified into seven types: labor, equipment, workspace, materials, precedence, information, and external flows (Koskela 1999). I tracked a total of 5,843 flows in this research, all of which fell into one of these seven categories. Flows released from upstream activities become inputs to downstream activities. Therefore, delays in upstream activities hinder the timely release of flows, which can cause delays in downstream activities depending on those flows. To manage the flows, field managers need to know the flows' source, their status, and their readiness likelihood. Current construction models do not formally represent, measure, and track all the flow types. Hence, field managers lack formal methods for tracking the flows' status and estimating their readiness likelihood. Instead, they rely on their intuition and experience to manage the flows. This dissertation presents an activity and flow-based construction model, called the Activity-Flow Model (AFM). The AFM allows field managers to proactively manage the on-site work by allowing them to formally represent, measure, and track the construction activities and flows. The AFM consists of an ontology that defines the representation of the activities, the flows, and their interactions; the planning and control methods that enable the AFM's implementation on site; and the predictive models that help anticipate variations in downstream activities. The AFM was developed based on literature, field observations, and feedback from field managers. The AFM was validated prospectively for a total of 26 weeks through its implementation on three building projects that were in different phases (foundations, core and shell, and finishing), locations (Bogota, Copenhagen, and Lima), and used different planning and control methods (master schedule and weekly planning, Last Planner System, and Location-based Management System). The AFM was able to represent all the activities (1,645) and flows (5,843) in the test projects, track their variations, and quantify their variability. The planning and control methods enabled field managers to proactively manage the projects taking both the activities and flows into account. The predictive models supported by the AFM allowed field managers to anticipate variations in downstream activities and outperformed the predictive models supported by the Resource-constrained Critical Path Method (RCPM) (Fondahl 1961) and Location-based Management System (LBMS) (Kenley and Seppänen 2009) representations. The field managers used the analytics of the activities' and flows' performance record to allocate resources, size buffers, and modify the look-ahead schedule. Hence, the AFM can help field managers improve flow readiness, reduce activity delays, and improve schedule conformance.

Description

Type of resource text
Form electronic; electronic resource; remote
Extent 1 online resource.
Publication date 2017
Issuance monographic
Language English

Creators/Contributors

Associated with Garcia-Lopez, Nelly P
Associated with Stanford University, Civil & Environmental Engineering Department.
Primary advisor Fischer, Martin, 1960 July 11-
Thesis advisor Fischer, Martin, 1960 July 11-
Thesis advisor Alarcón, Luis F
Thesis advisor Levitt, Raymond E
Advisor Alarcón, Luis F
Advisor Levitt, Raymond E

Subjects

Genre Theses

Bibliographic information

Statement of responsibility Nelly P. Garcia-Lopez.
Note Submitted to the Department of Civil and Environmental Engineering.
Thesis Thesis (Ph.D.)--Stanford University, 2017.
Location electronic resource

Access conditions

Copyright
© 2017 by Nelly Paola Garcia Lopez
License
This work is licensed under a Creative Commons Attribution Non Commercial Share Alike 3.0 Unported license (CC BY-NC-SA).

Also listed in

Loading usage metrics...