Loading

FLAPP Model

A novel accident causation model for construction systems
  • RESEARCH
  • SYSTEM SAFETY | ACCIDENT MODELING | CONSTRUCTION
  • VRSS Lab | School of Aeronautics & Astronautics | Purdue University

Failures in construction systems due to embedded pathogens

Buildings, bridges, and other civil engineering structures on the ground that are shaping our built environment (Bartuska, 2011) are all products of construction—making or forming “by combining or arranging parts or elements” (Merriam-Webster)—which can be described as built systems (Hollnagel, 2014, p. 223) or constructed systems. Within the scope of this research, extraterrestrial constructed systems refer to extraterrestrial habitats—enclosed structures capable of accommodating human occupants either on the orbit or surface of other celestial bodies outside of Earth’s atmosphere, and their supporting infrastructure with capabilities including communication, energy, and transportation. In this context, a construction system can be defined as an organizational and technological system that produces constructed systems by fabricating and installing materials into an integrated structure according to design specifications.
Failures—non-performance or inability of a component (or system) to perform its intended function (Leveson, 2011)—in construction systems—processes of designing and building constructed systems—can lead to various consequences, including structural failures, dysfunctional building systems, and prohibitive cost for maintenance. One notion of such failures is that latent failures in one or more defective processes in the project materialize in the constructed system and remain unnoticed until eventually breaking out as an accident. Borrowing the analogy of resident pathogens based on an epidemiological notion of failures (Reason, 1990), we can describe latent failures as embedded pathogens carried in the physical artifact.
Looking upwards, construction in space will require renewed attention to failures due to embedded pathogens, as addressing them becomes exceedingly difficult and costly after launch, and it is easy to imagine even failures with relatively minor consequences on Earth leading to severe or catastrophic consequences if they occur in space. In such a treacherous environment, leveraging frameworks and methodologies for risk assessment and safety management becomes crucial for paving the way forward, as consequential failures could halt or even reverse the entire endeavor. Hence the research question: how might embedded pathogens end up in extraterrestrial constructed systems, and how can we prevent their outbreak in space?

Shortcomings of existing accident causation models

In the context of my research on failures and accidents in construction systems, the time dimension is critical for capturing the propagation and persistence of accident pathogens. Construction projects involve dynamic networks of multiple organizations, where defects can become embedded in artifacts and persist through recurring processes. Both permanent industry-wide standards and temporary project-specific specifications are used for quality control.
A review of accident causation models in the literature reveals limited explicit representation of the time dimension, while confirming the increasing popularity of hierarchical structures—a crucial feature for capturing complex interactions in sociotechnical systems. The Unseamed Model (translation of ほころびモデル [hokorobi model] in Koma et al. 2008), which depicts both time and levels as separate dimensions within the construction context, aligns with my research scope. However, its lack of well-defined graphical grammar and vocabulary indicates room for improvement. Therefore, there is a need to develop a new accident causation model that incorporates both time and hierarchical levels as fundamental concepts for illustrating and understanding construction system failures.

Basic concepts of the FLAPP model

Following a comprehensive literature review, this study developed an accident model termed the framed-and-layered accident pathogen propagation (FLAPP) model. The basic concept of the model consists of frames representing the temporal dimension, layers depicting the hierarchical aspects of the sociotechnical framework of construction projects, and graphical notation illustrating the sequence of defective processes and pathogens embedded in the constructed artifact.
The figure below shows a matrix representation of the FLAPP model, using the Sasago Tunnel ceiling collapse accident. Each column represents a frame that belongs to one of the project phases, with the red shades corresponding to the different types of defective processes. The frames are ordered and numbered according to the start of the process in each frame, as noted in the bottom row of the matrix. The rows represent the layers: process definitions, player roles arranged in hierarchical and chronological order, physical processes, the artifact, and the physical environment. Here, a process definition refers to the policies, regulations, standards, and guidelines established by the regulatory and organizational control that transcends a single project, which defines how a particular process is to be carried out, such as in design codes, quality assurance policies, and inspection guidelines. The players executing the information process are represented by circles, and the connecting lines depict the input/output relations. The bottom three layers (physical process, artifact, and physical environment) remain blank until any physical process appears in the project. The red ovals in the artifact layer represent the embedded pathogen, which was the defective anchor bolts in the Sasago Tunnel case that remained unnoticed throughout the operation phase.

FLAPP MATRIX | Sasago Tunnel ceiling collapse
GRAPHICAL ELEMENTS OF A FLAPP MATRIX

The level of detail is adjustable—if we want to zoom into a specific frame, we can create a Frame view, which can describe all the additional details of each layer and specify the input and output to and from other frames, along with what is communicated between the layers. This frame view is one way to zoom into the details of the matrix representation, and we can also zoom out from the matrix to a schematic view that only shows the high-level overview of the involvement of players across the project timeline.

INDIVIDUAL FRAME VIEW | Details of involved players, defective processes in an individual Frame (Frame 6)
SCHEMATIC VIEW | High-level overview of player involvement across project timeline

References

  • Bartuska, T. J. (2007) The Built Environment: Definition and Scope. In The Built Environment: A Collaborative Inquiry Into Design and Planning (pp. 3–13).
  • Hollnagel, E. (2014) Resilience engineering and the built environment. Building Research and Information, 42(2), 221–228.
  • Koma, K., Furusaka, S., Kaneta, T., Hirano, Y., & Egashira, T. (2008) Hinshitsu-jiko-jirei kara miru kenchiku-seisan-shisutemu no jittai to sono zeijakusei [The fragility in the building delivery system through the construction failures]. Journal of Architecture and Planning (Transactions of AIJ), 73(623), 183–190.
  • Leveson, N. G. (2011) Engineering a Safer World. book. The MIT Press.
  • Reason, J. (1990) The contribution of latent human failures to the breakdown of complex systems. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 327(1241), 475–484.
Related Publications

Category: Research
Share: