BUILDING SIMULATION FOR CRIME PREVENTION IN A
VIRTUAL
ENVIRONMENT
Due to the alarming increase in the frequency of terrorist acts involving explosives, and the arguable failure of existing methods to contain this problem, there is a need for new tools and techniques to reduce the number and scale of such attacks. This project investigates and evaluates a building in a virtual environment to demonstrate the applicability and adaptability of Crime Prevention Through Environmental Design (CPTED) and other related principles for the purposes of establishing features that may be incorporated into the design and construction of crime-resistant structures. The use of virtual environments permits the inspection of structures with a depth of simulated realism, integrated into an interactive environment, not possible with traditional media devices. Another advantage of using virtual environments is that it allows the examination of structures not yet, or no longer, in existence.
On April 19, 1995, a bomb, which was made out of an estimated 4,000 lb. (1810 kg) of ammonium nitrate soaked in fuel oil, was placed in a truck parked outside the Alfred P. Murrah Federal Building in Oklahoma City and was detonated with a single stick of TNT. The resulting explosion brought down about a third of the nine-story reinforced concrete building, and more significantly, caused the deaths of 168 people, and injuries to more than 400 others [ENR 1995]. Two years earlier, the World Trade Center in New York suffered in a similar terrorist act of bombing that left six dead, and over 1,000 injured. Other significant incidents in the last two years include the bombings of the Saudi Arabian National Guard Headquarters in Riyadh, Saudi Arabia in 1995 (seven dead, 42 injured), and the Jewish Cultural Center in Buenos Aires, Argentina in 1994 (100 dead, 200 injured).
In each of these incidents, the bombings caused the failure of a building. It is not improbable to assume that if the building had been able to withstand the bomb-blast, the casualty figures would have been significantly smaller. In the case of the Alfred P. Murrah Building bombing, it as been estimated that up to 80% of the deaths were caused by the building collapse rather than the blast itself ICE 19951. While it is not feasible for civil engineers to design and build a building structurally so that it would be able to withstand the impact of a 4,000 lb. truck bomb, it might be possible for a building to be designed in a way that would reduce the damage caused by the impact of a bomb on the structure, as well as the likelihood of it being a bombing target by manipulating the environment in and around the building. Although there are recommendations set by the Department of Justice (DOJ) for security measures for high-risk buildings such as perimeter monitoring via closed-circuit television (CCTV) and installation of shatterproof glass on all exterior windows, they are often expensive to implement and subsequently, most federal buildings fail to meet the set standards [CE 1995].
A three-dimensional (3-D) model of a building to be constructed, or a model of an existing building with modifications, in a virtual reality (VR) simulation would allow experts to determine the effectiveness of the security measures that are to be implemented before the actual construction of a new building, or the refurbishment of an existing one. Potentially costly measures that are ineffective could then be identified at the planning and design stage, thus reducing the costs for such measures. A "virtual walkthrough" of the building in a virtual environment would also allow an expert to identify security measures not previously thought of, implement them in the 3-D model, and evaluate their effectiveness. A virtual environment simulation is able to do this as it affords the user the closest experience to actually "being there" through a multi-sensory media environment. Its simultaneous immersive and interactive qualities make it an ideal visualization platform, allowing the user to explore a building without actually having to visit the building.
In 1992, at the request of several federal agencies, the Committee on Virtual Reality Research and Development was established by the National Research Council [Durlach and Mavor 1995]. Since then, research and development in VR technology has produced a number of models for various applications. Many other simulations are being developed. Here at OSU, research in the use of VR has produced a number of simulation models for education and training purposes. An example of a current project is a virtual reality modeling of the construction of a highway overpass in New Albany, Ohio [Tsay and Hadipriono 1995].
GOAL AND OBJECTIVES
The goal of this study is to demonstrate the applicability and adaptability of Crime Prevention Through Environmental Design (CPTED) principles and other related principles for the purpose of establishing features that may be incorporated into building designs towards the evolution of bomb-resistant structures with a virtual environment simulation program. This program would allow users to inspect a building in a virtual environment to identify CPTED concepts that could be adapted for security measures, as well as allowing users to evaluate those measures after the modifications have been implemented. As such, the objective of this study would be to develop two modules: (1) An interactive virtual environment simulation of the original building, and (2) an interactive simulation of the building with the implemented security measures and features.
METHODS OF PROCEDURE
The seven major tasks in the development of this program were: (1) a review of relevant literature to acquire the required background knowledge; (2) the selection of the Alfred P. Murrah Building as the subject of the study, (3) the development of a 3-D model of the building based on the original building plans using Computer Aided Design (CAD) based packages; (4) the development of an interactive and immersive virtual environment by writing computer programs utilizing VR packages to incorporate the intrinsic coordinates of the created 3-D models into the global coordinate system of a virtual environment, and establishing the viewer's coordinate system (to track the movement of the user's line of sight) in the environment, and the development of interactive action controllers to control movement in the virtual environment using a cyberglove or a mouse for realistic control; (5) the investigation of the 3-D model of the building in virtual reality to identify potential changes based on CPTED and other related principles; (6) the modification of the original 3D model based on the identified changes; and (7) the evaluation of the modified design in the virtual environment by faculty members and graduate students of the Civil Engineering and Landscape Architecture departments at the Ohio State University.
The software packages utilized for the development of this program were Autodesk' s 3D Studio, a CAD-based graphical package which runs on the MS-DOS platform for the modeling of 3-D objects, and Sense8's WorldToolKit (WTK), a development system consisting of a library of C-language functions that serve as a set of tools for building 3-D virtual environments. The 3-D modeling was performed on a Pentium processor based personal computer (PC), while the virtual environment was developed and run on a Silicon Graphics (SGI) Onyx RealityEngine2 (Onyx RE2), a graphic supercomputer. Additional equipment used for navigation and interaction in the virtual environment included a cyberglove and a mouse for tracking in the virtual environment, and a head-mounted display (HMD) for full visual immersion in VR.
A BRIEF INTRODUCTION TO CPTED
CPTED is based on the assumption that "the proper design and effective use of the built environment can lead to a reduction in the fear of crime and the incidence of crime," [Jeffrey 1977]. Pioneers in the field of the use of the physical environment for crime prevention, such as Jeffrey, Newman and Jacobs, performed studies that eventually would form the core of CPTED over 25 years ago. Despite efforts by many government institutions, criminologists and security design experts since then, the concept of CPTED has yet to be accepted, and incorporated into projects by the majority of architects and engineers [ENR 1995]. Even so, townships and communities that have adopted CPTED concepts into their crime prevention services, such as the Region of Peel, Ontario, and Portland, Oregon, have reported significant reductions in the level of criminal activity after their implementation.
The three primary focuses of CPTED are surveillance, access control, and territoriality. These concepts may be applied to both macroenvironments, such as campuses and cities, and microenvironments like an office. Departing from the traditional method of crime prevention by target hardening where techniques such as fortressing, and using locks, alarms, or security guards, CPTED fights crime by examining, and manipulating the spatial configuration and the circulation patterns of the people using the space. This is achieved by examining the physical, social, and psychological needs of the expected users of the space, the intended use or non-use of the space for the expected activities, and by predicting the behaviors of the expected users as well as the offenders [Crowe 1991].
Surveillance, by definition, means to keep a close watch over someone or something. In CPTED, surveillance is used as a design concept for the purpose of keeping potential intruders or criminals under constant observation [Crowe l991]. Thus, surveillance strategies should facilitate the ease of observation of the environment, and these strategies may be classified as one of three types- (l) organized (e.g. security patrols); (2) mechanical (e.g. lighting); and (3) natural (e.g. windows). Although CPTED primarily emphasizes natural forms of surveillance, it would be prudent not to neglect the other forms of surveillance.
Access control, in CPTED, is another design concept with the objective of decreasing criminal activity by denying access to a crime target, and by creating a perception of greater risk in potential offenders. As with surveillance, access control strategies may be divided into three categories: (1) organized (e.g. security guards); (2) mechanical (e.g. locks); and (3) natural (e.g. spatial definitions). Natural access control is achieved by first classifying areas in the environment into zones that fall into one of three categories: (1) public, (2) semipublic, and (3) private. Boundaries between zones are then established by the use of barriers. Barriers may be physical (e.g. walls or partitions) or non-physical (e.g. different carpet colors or floor-tile patterns).
Recent approaches to crime prevention by incorporating a greater use of natural techniques such as natural access controls and surveillance, have resulted in the emergence of the territoriality concept. The concept of territoriality strives to instill a sense of proprietorship among the users of a space through the creation of an extended sphere of influence in the physical design. This increased sense of proprietorship would, in theory, promote more responsiveness among the users in the protection of their 'territory', and increase the perceived risk in the eyes of a potential offender.
THE ALFRED P. MURRAH BUILDING
Completed in 1977, the Alfred P. Murrah Federal Building was a nine-story reinforced concrete building owned by the General Services Administration (GSA). Tenants at the facility included 17 federal agencies and defense-department offices such as the Social Security Administration, the Drug Enforcement Administration, Bureau of Alcohol, Tobacco and Firearms, and the Marine Corps, as well as a government-operated day-care center [Newsweek 1995]. At the time of the incident, an estimated 453 employees, 24 day-care center children and an undetermined number of visitors were in the building.
According to a DOJ study of federal buildings conducted after the bombing incident, the Alfred P. Murrah had a security classification of Level 4 (facilities such as the Pentagon or CIA headquarters are Level 5 structures). The DOJ study recommend further security measures for buildings in Level 4 or 5 such as "control over facility parking, perimeter monitoring via closed-circuit television (CCTV), intrusion detection systems, x-ray screening of all incoming mail, installation of shatterproof glass on exterior windows and a setback from the street for new buildings," [CE 1995]. It has been estimated that, to implement these features into a new building, an additional $2.5 million in construction costs would be incurred, and that it would take $3 million to completely refurbish an existing structure.
THE BOMBING AFTERMATH
Structural experts agree that the placement of the truck-bomb, approximately 15 feet (4.6 m) from the front of the main entrance of the Alfred P. Murrah Building, was, from the perspective of a terrorist, the ideal location to cause the greatest damage to the building. In addition to creating a crater measuring 30-feet (9 m) wide and eight feet (2.4 m) deep, the force of the explosion also ruptured three of the four exposed supporting columns in the twostory entrance foyer. The failure of these columns resulted in a progressive collapse of the floor slabs, breaking off from the center columns, all the way to the roof of the building. This mode of failure has been likened to the knocking-off of a single card at the base of a houseof-cards.
Constructed over 20 years ago, the design of the Alfred P. Murrah Building did not take into account the threat of a bomb. Even today, high-risk buildings are not designed to withstand the impact of a 4,000 lb. bomb placed 5 yards away. A GSA panel formed three months after the bombing concurred with the recommendations of the DOJ, and suggested that a careful examination of the design and construction process was required, but the primary issue of the discussions was the use of controlled accessibility "to prevent intrusion and crimes against people and property," [CE 1995]. Eve Hinman of Failure Associates, Inc., Menlo Park, California, a panelist, stated that, due to the high costs of strengthening existing buildings, it would be more prudent to shift the emphasis onto non-structural security.
THE 3-D MODELING OF THE ALFRED P. MURRAH BUILDING
The blueprints for the 3-D modeling of the Murrah Building were obtained from the General Services Administration, and it needs to be noted that the blueprints are for the original design, and any renovations that may have been done on the structure after its completion in 1977 are not be reflected in the model. The model, developed as 58 separate objects consisting of approximately 60,000 polygons stored in a file approximately 1.55 megabytes in size, represents the external structure of the building, and the general properties of the roads directly around the building. Below (Figure 1), is a 3D Studio rendered image of the model of the Murrah Building.
Figure 1: The front-view of a 3-D Studio rendered image of the Alfred P. Murrah Building.
CREATING VIRTUAL ENVIRONMENTS WITH WTK
WTK is a development system consisting of a library of C functions that serve as a set of tools for building 3-D virtual environments. These functions, when implemented in a program, manage tasks such as importing model geometries, rendering, and reading input sensors. The core of a WTK-driven application is comprised of simulation loops that continuously read the input sensors, update the 3-D objects, and render the new view of the universe. Although WTK is platform independent, the size and complexity of the models in this project are such that it requires the computational and graphical capabilities that only a graphic supercomputer such as Silicon Graphics' Onyx RealityEngine2 can provide.
The first step in developing a virtual environment with WTK is the creation of a "universe", essentially a container for all the 3-D objects. The universe needs to be initialized as well as configured to the output display by which the universe is to be viewed. Conversely, at the end of the application, it needs to be terminated to free up all the memory that has been used during the progress of the application. Once a universe has been created, it can then be customized by defining the properties of the universe such as the size of the viewport, the initial location of the viewpoint, and whatever hardware the application may utilize (e.g. keyboard, mouse, cyberglove). The next stage of developing a virtual environment builds on the previously defined universe by loading all the static 3-D objects in the model, including light sources. Although it is still possible to load objects while the application is in its simulation loop, it is much more efficient to load all static objects before the loop. Once the universe has been set up and all the static objects loaded, the application goes into its continuous simulation loop of reading the input sensors, updating the 3-D objects, and rendering the new view of the universe.
In addition to the standard library of functions provided by WTK, there are numerous supplemental libraries available. These libraries are usually created by the users of WTK, and expands on the standard library by providing additional functions that supplement the standard functions. Most of these additional libraries, or add-ons, are available over the Internet as freeware or shareware. Unlike the standard WTK library, not all the add-ons available are platform independent, nor are they guaranteed to be error-free. AddOns for WorldToolKit (AWT) is such a library. Created by Tsung-chieh Tsay, a former graduate student of the civil engineering program at the Ohio State University, it consists of six primary functions that were initially developed for the project Construction Operations in Virtual Reality (COVR) [Tsay and Hadipriono 1995]. This library provides features such as multiple universe actions, multiple object tasking, collision detection and non-penetration of objects and viewpoints, and enhanced text and button capabilities.
INTEGRATING OF THE 3-D MODEL INTO THE VIRTUAL ENVIRONMENT
In the porting of the 3D Studio file of the Murrah Building model to the virtual environment created with WTK, it was discovered that the resulting 3-D model had been altered in terms of the color and textures of the objects. Textures that were custom-created in 3D Studio did not appear at all in the ported model. The reason for this was the inability of WTK, running on the Onyx RE2, to read texture files in formats other than SGI's natural image file format. While it is possible to convert these 3-D Studio created texture files, it was decided that, to maintain the frame-per-second (FPS) rate, the 3-D model used in the virtual environment should contain a minimal amount of texturing. The FPS rate is important as too low a rate could cause a phenomenon known as cyber-sickness. The FPS rate is limited by the processing power of a computer, and with current technologies, a trade-off is required between the level of detail in a model and the FPS. These problems notwithstanding, the model of the Murrah Building was successfully loaded into the virtual environment, and the investigation of the original design was performed.
THE INVESTIGATION OF THE MURRAH BUILDING
The investigative "walkthroughs" of the Murrah Building in a virtual environment revealed that there were deficiencies in the original design that could have been overcome by using the CPTED approach in the design process. Based on the above-mentioned investigations, and evaluation by experts, the following are a list of recommended changes and implementations that could have been applied to the building to reduce its vulnerability to bombings. It needs to be noted that not all of the following recommendations are based on CPTED principles, but in order for a security scheme to be effective for high-risk structures such as this building, other security concepts need to be incorporated.
CONCLUSIONS
The investigation of the Alfred P. Murrah Building has revealed that there were many steps that could have been taken to reduce the extent of the damage, both to property and people, caused by the bombing incident. Based on the feedback of the participants in the evaluation process, this research demonstrates that there were deficiencies in the original design of the building that could have been identified and thus overcome by using the process outlined in this paper during the design stages. Although there were minor problems in implementing the virtual reality concept, such as texturing and FPS rate, we expect that, with the rapid advances in computer processing power and parallel processing, the use of virtual reality and virtual environments will become a powerful and efficient tool for the evaluation and investigation of structural designs, perhaps even becoming a standard and integral part of the design process in the future. We believe that the program developed in this research will contribute towards increasing the awareness as well as providing a better understanding regarding the potential use of CPTED and its related concepts in building design among architectural and engineering students.
This work was supported by the Engineering Experiment Station (EES) of the Ohio State University. We would like to thank Prof. B. Breeden for his valuable comments on this project.
ENR (1995). "Oklahoma Blast Forces Unsettling Design Questions," Engineering NewsRecord May 1, 1995, McGraw-Hill Companies, New York, NY, 10108.
CE (1995). "Oklahoma City Aftermath," Civil Engineering October 1995, American Society of Civil Engineers, New York, NY, 10017.
Newsweek (1995). "Get Me Out of Here," Newsweek May 1, 1995, Newsweek, Inc., Livingston, NJ 07039.
Durlach, N.I. and Mavor, A.S. (1995). Virtual Reality Scientific and Technological Challenges, Sponsored by National Research Council, National Academic Press, Washington, DC.
Tsay, J. and Hadipriono, F. C. (1995). "Virtual Reality Modeling for Bridge Construction." The 1996 American Society of Civil Engineers Third Congress on Computing in Civil Engineering, Anaheim, California (accepted).
Ray, Jeffery C. (1977). Crime Prevention Through Environmental Design, Sage Publications, Beverly Hills, California.
Back to Table of
Contents