Virtual (3D) Collaborative Environments: An Improved Environment for Integrated Product Team Interaction?

Transcription

1 Virtual (3D) Collaborative Environments: An Improved Environment for Integrated Product Team Interaction? Grace M. Bochenek Director, Advance Virtual Environments Lab U.S. Army Tank-automotive and Armaments Command (TACOM) Warren, MI James M. Ragusa Associate Professor University of Central Florida Department of Industrial Engineering and Management Systems Orlando, FL Abstract To shrink the cost and protracted length of total system and product development life cycles (DLCs), many organizations (including the U.S. Army) have moved away from serial to concurrent collaboration through the use of cross-functional, integrated product teams (IPTs). In addition and more recently, a growing number of these organizations are also using life size three-dimensional (3D) virtual collaborative environments (VCEs). This enabling technology provides individuals and IPTs with views almost as realistic as physical prototypes, and offers the capability to rapidly change perspectives to view the outside, inside, top, and underside of synthetic system and product model(s)--even full scale. This paper addresses and provides an answer to a central research question: Do 3D VCEs provide an improved environment for IPT team interaction? The answer is: Yes, they do! 1. Introduction To overcome the high cost and duration of military weapon and support system life cycle times and the limitations of traditional system acquisition practices, the U.S. Department of Defense (DoD) has developed and adopted a strategic technology and innovation management initiative called Simulation Based Acquisition (SBA) to support the transformation needed to create the military of the future. This integrative process is designed to promote the rapid and more economical development of quality military systems through the use of computer-based modeling and simulation and integrated project teams (IPTs). Goals include the reduction of all future total system acquisition cycle times by 50%, and the adoption of SBA by all service branches [1]. The overall objectives of SBA are: (a) substantially reduce the time, resources, and risk associated with the entire acquisition process, (b) increase the quality, utility, and supportability of fielded systems while reducing total costs throughout the total life cycle, and (c) enable the use of IPTs across the entire acquisition life cycle [1]. Implementation objectives of this initiative are to encourage and support iterative, cross-functional team collaboration, and early user system evaluations. [1] In a similar vein, the U.S. Army has launched the Simulation and Modeling in Acquisition, Requirements, and Training (SMART) Program to fully include and integrate all of its stakeholders (users, program managers, contractors, testers, trainers, maintainers, and logisticians) into the Army s modernization efforts [2]. Important players in SMART implementation are the Warren, Michigan based Army Tank-automotive and Armaments Command s (TACOM) Tank-Automotive Research, Development, and Engineering Center (TARDEC) and its National Automotive Center (NAC). Members of the Army community are now regularly using virtual collaborative environments (VCEs) for system and product development life-cycle (DLC) activities as a way of implementing SBA and SMART directives. These enabling information technologies are based on virtual reality (VR) and three-dimensional (3D) 1-to1 displays. Included are 3D computer graphic systems with real-time user interactive control and viewer-centered perspectives that provide IPT decision makers with a collaboration and solution environment during the DLC. In this way, program and project management, engineering, and system users are concurrently involved in a full range of system requirements generation, design, development, and operational use evaluations.

2 Profit-making, private sector organizations are also using VCE tools, with similar improvement goals in mind, to improve their product design, development, and manufacturing processes [3,4]. These applications have evolved from earlier developments in using simulation and modeling and now VCEs for industrial design and development activities [4]. Examples include: Boeing for its 777 aircraft [5], DaimlerChrysler for its Dodge Intrepid product line and engine development [5], General Motors for reviewing vehicle concepts in various outdoor seasonal settings and comparisons to current models [6,7,8]. These organizations and others report significant cost avoidance and DLC time-reductions when compared to more traditional IPT collaborative review and development methods and activities. Legacy examples include: using and reviewing paper drawings, computeraided design (CAD) generated drawings and images, PowerPoint TM presentations, physical prototypes, and computer-aided engineering (CAE) information. However, while these large corporations are using VCEs in prototyping and developing new products, little empirical testing has occurred [6] in the past. With some exceptions, the same is true to date. This paper describes: the research methodology used; a conceptual IPT/VCE model; VCE technologies and systems; commercial and military applications and empirical testing; operational findings and research issues; and a study summary. Answered in this latter section is the central research question asked in this paper: Do 3D VCEs provide an improved environment for IPT team interaction? 2. Research methodology The findings and conclusions presented in this case study are based on a four-year and continuing research and development effort involving the authors and others. Diverse discipline areas included in a literature review included: engineering and project management, individual and organizational behavior, IT, computer supported collaborative work (CSCW), and VR. The literature review and subsequent interviews with military and civilian subject matter experts resulted in the development of a conceptual model that includes various parameters and interrelationships important to program/project management and IPT members working in this new VCE. Empirical testing focusing on several model parameters was conducted, and applications and test results from other organizations reviewed. To validate the model and to collect additional information about operational factors and research issues, case study analysis and user in-depth interviews were conducted. Literature searches, empirical test data, and case study information were analyzed and integrated to determine if the developed conceptual IPT/VCE framework and its parameters and interrelations could be supported or refuted. Finally, operational findings and research issues were identified. 3. Conceptual model The conceptual IPT/VCE model, shown in Figure 1, (adapted and modified from [9] and [10]) illustrates the interrelationships between people, work, and technology within the broader context of organizational, social, and technical environments. Selecting an appropriate tool or integrating advanced technologies to support the tasks of team product design evaluation and development requires that all model elements be considered individually and collectively. While this can be considered a general model, this paper focuses on how these elements relate to and affect the use of VCEs by IPTs. Organizational Environment People Work Technical Environment Social Environment Technology Figure 1. A conceptual IPT/VCE model 3.1. Organizational, social, and technical environments There are several environments that drive and influence all activities and tasks within organizations and the DLC process. As shown in Figure 1, these environments are: organizational, social, and technical. The organizational environment is primarily defined by the goals and objectives of the organization--whether profit seeking or not. The social environment is determined by how the organization functions with regard to individual and team interaction, cohesiveness, and motivation to accomplish work. The structure of the organization, e.g., functional, project, or matrix, affects the social context of activities performed. The technical environment is driven by the degree to which the organization supports and is dependent on technology. An organization s degree of technical orientation and dependency can lie anywhere on a continuum between highly innovative and technology (risk) averse. These environments by themselves and jointly, influence a set of factors that directly affect IPT DLC

3 activities. As illustrated in Figure 1, these elements are work, people, and technology. Within these factors, important are the relationships between people and the work tasks they perform, and between people and the technology they use to accomplish work Work: product design and development In many contemporary organizations, product design and development are major objectives and reasons for the existence of certain organizational elements and IPTs. Ulrich and Eppinger [11] have characterized product creation as a systematic method of evolving a product from idea conception to product release to customers. The process begins with concept development, and progresses through system level design, detailed design, testing and refinement, and production ramp-up [11]. Current product development practices seek to reduce overall system DLC times through the use of concurrent engineering and multi-functional and empowered IPTs. Usually required are flexible leadership, frequent design iterations, simulations, and extensive testing. For many organizations resources used, development time and cost, time to market, and resultant product quality are used to measure the assessment of performance and success in product development. Two important metrics, overall product costs and time to market, are dominated by decisions made in the early stages of the design process. It has been estimated that 85% of product development costs are determined before the product design is released to manufacturing [4]. Others estimate that by the time 10% of total project funds are expended, approximately 90% of a product s development costs are established [12]. In other words, the most cost- and schedule-determining decisions in the product development effort occur early in the DLC--during the design phase. As should be evident, design errors detected and corrected during early design phases have the highest likelihood of reducing overall product or systems costs and time to market since it is easier to change electrons than atoms. During this early DLC phase, planning and design decisions are made to identify and reduce risk and improve total product quality. Because of the cost implications involved in early design activities, organizations need to select systems and methods that provide cross-functional design IPTs with the best decision support tools available People: integrated product/process teams Traditional methods of system and product design, development, and acquisition have often been described as serial and linear processes where functional groups make contributions sometimes independent of other internal and external elements [13]. As a result of these throw it over the wall activities, critical decisions that can significantly impact overall product design, development, and costs are frequently made without regard to the effect on or consequences to others. One of the basic ideas of concurrent engineering, needed for product design and development, is to assemble a team that is focused on developing or redesigning a product. These teams are usually composed of people from various functional elements, e.g., development, engineering, manufacturing and product management [14]. Concurrent engineering, is not a set of techniques but a conceptual methodology that enables all who are impacted by the product design have early access to design information and have the ability to influence the final design to identify and prevent future problems [15]. Within these cooperative environments it is necessary for individuals to share information and collaborate in the decision making process. Several researchers and authors indicate that collaborative work requires the exchange of information for purposes such as notification and clarification, and the processing of information for monitoring, negotiating, and decision-making [12, 16, 17]. This cooperation ensures that everyone impacted by the design has early access to design information and the ability to influence the final design effectively and efficiently. Collaboration is the key to make this happen. Unfortunately, in spite of the vast body of collaborative work data collected over the last half century, an underlying theory of collaboration does not presently exist--let alone one that relates to VCEs. For the above reasons, a need exists to develop tools that support concurrent engineering and cross-functional design team interactivity by providing individuals and teams with accessibility to design data and information regarding product versions and customer needs. To improve the probability of product development success, these tools should allow teams to interact face-to-face and make decisions from multiple perspectives in a shared information space, sometimes at distributed (networked) locations, using the best and most current information Technology: immersive, interactive VCE tools Frequently, there is a need to change views in a product or system development environment. A commonly used method includes manipulating Computer Aided Design (CAD) images. Within a CAD environment, views can be rotated, reduced, enlarged, or changed fairly easily. However, usually only reduced scale (scaled down) images are available in these systems. The VCE is different from traditional two-dimensional (2D) paper or 3D electronic CAD drawing review environments because it provides and supports users with full-scale (1-to-1) system or product views. Within a 3D VCE, numerous navigation and viewing perspectives (on

4 top of, beside, underneath, inside of) perspectives are also possible. These systems and their interface devices also provide a capability for 3D, real time image reorientation. Various VCE systems technologies and interface devices, discussed in the next section of this paper, are available to implement 3D and real time viewing of product models. Some have multisensory capabilities that support visual, audio (including directional sound), direct object manipulation, and touch. Important to VCE users are the sensations of immersion (being surrounded in an environment) and presence (feeling you are in one place while actually being in another). Both occur in some degree in all VCEs. With the IPT conceptual model and its parameters and interrelations in mind, the next section identifies and briefly discusses product design technologies and two commercially available VCE systems and their features. These systems, individually and together, are presently being used by several public and private sector organizations to support various phases of their DLC and IPT activities. 4. VCE technologies and systems 4.1. Product design technologies Computer technology has had a pervasive and significant impact on the total product design process. In the past, products were designed and developed using pencil and paper drawings that only represented 2D product views. Today, product concepts are routinely initiated by developing 2D and 3D solid models of alternative designs using computers. CAD technology enhances the productivity of a single designer, but is not very productive or effective in collaborative group design or development review settings. VCE, often-called virtual reality or VR, is a technology that significantly supports the needs of today s IPT DLC processes. A VCE is a suite of 3D graphics, simulation tools, and displays that allows users to operate within a computer-generated environment on an interactive real-time basis. In some organizations, VCEs are starting to be accepted as a general business tool. For example, in the commercial automotive industry, the use of traditional clay models and physical prototypes is costly. Using VCE systems, a team can visualize and interact with complete virtual prototypes consisting of assemblies, sub-assemblies, or components in 3D and explore new products, plans, and concepts long before they exist in reality. Participants meet in a virtual showroom setting to review a combination of styling, packaging, and engineering aspects of a proposed vehicle. In order to substitute for physical prototypes, virtual models must provide high-fidelity renditions of vehicles, contain physical and functional features like shininess (highlights, color, reflections), support contour changes (i.e. flexible parts), and a have the capability to open and close a virtual door VCE systems As part of a continuous improvement strategy, the Army and numerous commercial organizations have purchased advanced VCE systems that they feel better support their DLC time and cost reduction objectives and activities. The first system purchased by the Army was the Cave Automatic Virtual Environment (CAVE TM ) developed by the Electronic Visualization Laboratory of the University of Illinois-Chicago. The second was the semi-immersive WorkWall TM developed by Fakespace Systems Inc. Both have capabilities for use by individuals and IPTs for VCE-based reviews. These systems and their interface devices can be programmed to orient images in real time while review participants remain immersed in the viewing environment. General multi-sensory capabilities include visual, audio (including directional sound), direct manipulation, and touch. This mix accommodates all human senses except taste and smell, which are usually not very important in most product or system developments The CAVE. The CAVE system is an immersive, multi-person, 10x10x10-ft 3 room-sized, high-resolution 3D video and audio VCE. Images are created for the CAVE environment beginning with CAD engineering data. Though the use of special translation software, synchronized 3D views are projected and reflected by Mylar mirrors on the CAVE front and side transparent walls and floor. The 45 degree mirrors are used to reduce the volume required to house the overall system. Figure 2 illustrates the layout of a CAVE showing the four projectors, reflecting mirrors, and walls and floor setup. Figure 2. The CAVE projection/reflection system In operation, users enter the CAVE, stand or sit, and have view navigation and selection controlled by a single operator using either a 6-degrees of freedom headrepositioning device or a Wand (essentially a computer mouse shown hanging in the center of Figure 2). The

5 Wand, allows an operator (or user) to interact within the VCE and reach out, pickup, move, and relocate projected 3D virtual objects. Normally the number of participants in the CAVE is limited to the size of the enclosure--with a maximum space limitation of twelve but practically five. Pictured in Figure 3 is a soldier reviewing the inside of a new concept design in a CAVE while wearing a special glove that provides haptic (touch) control of virtual objects. For additional CAVE information see [18]. and are ideal for IPT presentations and collaborative reviews. Unlike the immersive CAVE, this semiimmersive system provides concurrent 3D viewing by a number of participants limited only by room size. Each of the above VCE systems uses large projectors and stereoscopic glasses that allow IPT members to individually or collectively view virtual product or system models. In this way multiple viewers share virtual experiences, discovery, and ideas. 5. Commercial and military applications 5.1 Commercial applications Figure 3. The CAVE environment Figure 4. A WorkWall TM collaborative review The WorkWall TM. Similar to the CAVE system, a WorkWall TM system by Fakespace Systems Inc. [19]) is pictured in Figure 4 (being used by a commercial aircraft review team). It is a rigid, flat vertical high-resolution 3D video and audio environment. Models or environments are displayed from floor to ceiling. Screen sizes range to 30 by 30 feet depending on room and screen size. The system provides completely seamless images projected from multiple video sources. Most applications of this VCE technology are full 1:1 scale reviews of system exteriors As mentioned earlier in this paper, profit-centered private sector organizations have been using VCE tools and IPTs to improve their product design, development, and manufacturing processes [3, 4]. These applications have evolved from earlier developments in using modeling and simulation for industrial design activities [4]. For example, Keller [5] reports that the Boeing Corporation used Computer-Aided Three-dimensional Interactive Application (CATIA) software to reduce by 60 to 90 percent design rework for its 777 aircraft [5]. Various auto manufactures have also experienced substantial improvements and acceleration in vehicle styling and design made possible through the use of VCEs [17]. For example, DaimlerChrysler Corporation s Dodge Intrepid product line was developed using what the company calls cyber synthesis that resulted in five new vehicles and three new V-6 engines. As a result of the simulation of digital models, the company has reported cost savings of $75 million and a 20 percent reduction in its Intrepid model development time [5]. In addition, General Motors uses VCE system to review division styling concepts [7, 8]. In most of these applications, significant cost avoidance, reduced development time, and participant satisfaction have been reported. For more on the commercial use of VCEs see [4, 12, 20, 21, 22] Military testing and applications Empirical testing and results. The success of several commercial and early Army applications led the authors to the realization that additional user testing and reviews of VCE capabilities were needed. With the Figure 1 conceptual model as a framework and a specific VCE technology (the CAVE) in mind, empirical testing was conducted to provide insights and answers to some operational questions and to identify research issues. An initial experiment was conducted at the U.S. Army Mounted Warfare Test Bed Facility located in Ft. Knox, Kentucky. In an effort to improve Army conceptual design review processes and to reduce overall cost and time to become operational (e.g., time to market), the traditional method of new concept presentation (a stand-

6 up briefing using PPT slides) was compared to the use of a briefing conducted in a CAVE. The design review example chosen for the test was the conceptual design of a futuristic combat vehicle. Test participants were made up of 40 subjects consisting of regular and reserve soldiers grouped into eight-five person teams. None had participated in a VCE design review before. For more on this testing see [23] and [24]. Analysis of test date determined that immersion correlated positively with presence, and the use of CAVE as a tool for design review was favored. Importantly, however, it was found from qualitative data analysis that test participant resoundingly preferred the CAVE to the PPT concept presentation. Participants indicated that the CAVE provided significantly higher visualization (e.g., naturalness, real-world, responsiveness of movement), and greatly improved comparative values (e.g., degree of holding interest, quality of visualization, improved concept comprehension) than did the PPT format. Some participants were even observed trying to reach out the touch the virtual model [23, 24]. A post-test discussion involving test subjects also indicated significant advantages of CAVE functionally. Test participants noted that the CAVE immersive environment allowed them to move around and within the full-scale vehicle model and to get a far more real and robust sense of the concept. The great majority indicated strong support for the CAVE as a future concept review environment. One participant commented: The CAVE allowed me to really see and better understand the design I could touch and feel it around me. Another said, I can t wait to come back and see and react again to this scenario, and This is where the future is [23]. Future work continues to provide a better understanding of design comprehension and its relationship to immersion and presence. For more information see [24] Operational applications and some results. The CAVE and WorkWall are now routinely used by Army IPTs at different phases of their system life cycles. These VCEs provide a synergistic means to develop methods and systems that can be used to integrate advanced computer visualization and simulation technologies to overcome some of the limitations of traditional, costly, and time consuming, system design, development, and acquisition activities. Over the last several years various weapon and support system designs have been compared, evolved, and finalized by Army and contractor IPTs using CAVE and WorkWall VCEs. Rather than participate in a traditional design review viewing drawings, briefing charts, and text descriptions, these participants were immersed around, on top of, and inside design alternatives and can even observe dynamic system functional behaviors. Most VCE users have come to realize that months will be reduced from phases of the DLC process and probably years from the overall process. IPT members (including program and product managers, system users, engineers, trainers, testers, logisticians, maintainers, etc.) participate depending on the phase and objectives of the review. These members jointly evaluate system issues, ideas, parameters, and performance--each from their own experience, perspective, viewpoint, and functional responsibility. This unique multi-functional virtual system development environment has allowed all members to share in a concurrent experience and has given them to opportunity to simultaneously relate to the perspectives of others. The improved IPT communications facilitated in the VCEs significantly reduces the time to reach a common understanding and come to a consensus during evaluations of proposed alternatives and designs. To illustrate the benefit of an immersive VCE, an Army tank commander and other crewmembers (the customers) are placed into the proposed design of a future vehicle to visualize and experience its interior and ergonomic design. Invaluable observations from these eventual operators are then used to concurrently identify problems, address issues, and exchange ideas in real-time collaboration with other IPT members. The use of VCEs for product development provides design flexibility by allowing the exploration of various options and the opportunity to generate and iterate what if exercises early in the design process where mistakes or changes are inexpensive to correct. In this way, significant cost savings can be achieved by IPTs because numerous problems can be identified and corrected prior to actual physical product construction. Some user observations and comments obtained during operational VCE Army reviews are as follows: Director, Combat Development Office, Seeing a draft requirement function within an operational environment is much better than a large chart presentation. A system developer stated: Reviewing the designs in the CAVE with the engineers discussing characteristics of the subcomponents allowed me to very quickly compare my requirements to the concept design capability. I am interacting with design, engineers and staff simultaneously. Things become more informal and we quickly get down to business in our trade-off analysis. A Chief Engineer for the Office of the Product Manager said: It (the CAVE) gives us the opportunity to visualize functionality of concepts when reviewing engineering change proposals. Another Product Manager stated, Yes, seeing the designs and their movements helped speed up the decision making process [25]. The success and operational uses for these early VCE applications and reactions have fueled considerable recognition of the importance, interest, and demand for its use in many more ongoing Army programs.

7 6. Operational findings and research issues The following section identifies and briefly describes some operational findings (i.e. success factors) and research issues relating to IPT DLC activities in VCEs derived from a review of relevant literature, empirical testing, and case study data and information. The framework for analysis and development was the conceptual IPT/VCE model described earlier (see Figure 1 in Section 3). The elements of this model were reclassified from: organizational, social, and technical environments; and work, people, and technology categories. The new categories are operational and research with further division into individuals, teams, and system subcategories for each. In this way IPT/VCE operational findings are perhaps a little more useful to practitioners, and research issues more interesting to academicians. The following codes are included to indicate the fields of academic research and study that have been used and are likely to provide further insights. They are: E/PM = Engineering/Project Management, I/OB = Individual and Organizational Behavior, (IT) = Information Technology, CSCW = Computer Supported Collaborative Work, and VR = Virtual Reality. 6.1 Operational findings For the purpose of this discussion, operational findings are defined as those success factors that are associated with ensuring that IPT objectives and VCE system-based activities are more effectively and efficiently accomplished. We believe that once recognized and addressed, the following factors can be integrated into the DLC process to enhance IPT and participant success. Because of the summary nature of this paper, complete references are not included for more commonly accepted understandings Individuals Qualifying/testing participants. Participants must be capable of performing in a VCE. Testing (written and physical) may be required depending on the purpose of the review and the system used. Examples of testing include: susceptibility to cybersickness, color blindness, and adequate mobile skills. Cybersickness, or VR-induced symptoms and effects (VRISE) as defined by Cobb, et al [26], although not serious and short lived, can be irritating for many and more serious for some. According to Slater, et al [27] symptoms of VRISE may include headaches, things being out of focus, and feeling sick and sweaty. Color blindness and mobility testing are also important if IPT activities require participant true or relative color determination or physical motion capabilities. (VR) (CSCW) (I/OB) Teams Team composition. It is important that teams be composed of one or more qualified representatives from each applicable cross-functional area. An operational issue is: (a) Should virtual collaborative design teams consist of people who already know how to use the virtual display system? or (b) Will teams be make up of new members who have needed skills and functional knowledge but are not familiar with VCE technology? The latter has been found to be the most likely case and more realistic because VCE familiarization occurs very quickly for most users [23]. (E/PM) (I/OB) (VR) Leadership responsibilities. Team leaders and facilitators must identify the focus and content of task activities. In addition, once the review is underway they must keep the team focused and directed at achieving development goals and objectives, maintaining schedules, and performing in a professional manner in the VCE. Occasionally, members become overwhelmed by the technology and become distracted from purpose. The leader/facilitator should always attempt to bring the team to a consensus rather than merely a majority position. Indications exist that this leader should be the CAVE navigator [27]. (E/PM) (I/OB) (VR) Social relations. It is important to establish and maintain good social relations between team members. This requires that members feel part of the team, are allowed to make contributions to the objectives of the activity, and are not threatened. Slater, et al [27] in studying participants working in virtual and real-world environments, discovered a positive relationship between presence--being in a place, and copresence--the sense of being with other people. In addition, accord in the group increased with: (a) presence, (b) the performance of the group, and (c) the presence of women in the group [27]. (E/PM) (I/OB) (VR) Change control configuration management. It is important that approved IPT design changes and decisions be tracked and recorded for present and subsequent reviews. In addition, a record of the rationale for these decisions should be kept. Without this control and record of rationale, subsequent groups may reverse previous agreements without the benefit of earlier understandings. Until automated systems are developed to maintain a record of agreed upon design changes (a research issue), a team member should be designated to maintain a record of all team-approved changes. (E/PM) (IT) (CSCW)

8 Systems Operational readiness. The VCE system should remain stable and ready for use without breakdowns and interruptions. All equipment should be working before the IPT review starts. In the event of a breakdown someone must be on standby to resolve operational problems or lack of capabilities. During down periods team members should be encouraged to talk among themselves. Several basic questions are: (a) Does the organization have sufficient technological resources to support and maintain highlevel VCE environments and devices? and (b) Will individuals and teams have regular and reliable access to required technology to support their task responsibilities? (E/PM) (IT) (CSCW) Visual environment system familiarization. IPT members must be familiar and comfortable with the VCE equipment used. Training and familiarization are mandatory and should be done before the start of the IPT session and with the actual (virtual) equipment to be used. For many team members, this may be their first exposure to the new design environment. Purschke, Schulze, and Zimmermann [28] believe that the cardinal reason for the missing acceptance of computer-aided tools in (automobile) styling is the lack of intuitive and creative methods and concepts for communicating with the computer-- without technical abstraction. That is, without being blocked by complicated procedures of computer interaction. (E/PM) (CSCW) (VR) (IT) Navigation control. Navigational control should be given and maintained by a designated team member. This may be the team leader or a member selected because of special needs or skills. In VCEs, it is possible to take turns having control of objects (and perspectives) using what is know as baton passing. That is, the one who holds the baton (wand in a CAVE environment) is in control of negotiations. As noted earlier, some research indicates that the leader should be the navigator [27]. (E/PM) (CSCW) (VR) Networking remote sites. Because of the potential for distributed VCE design reviews at locations remote from a host site, compatible VCEs and networked capabilities must exist to support team activities with high-speed computer interfacing over extended distances. Since sites can be located anywhere nationally and internationally, direct networked high data rate communications (virtual image control, voice, and data) bandwidths and physical space will be required. Under certain circumstances there will be a requirement for VCE and network system security to protect classified information and proprietary product features. (IT) (CSCW) (VR) 6.2. Research issues Research issues are defined as those IPT/management and technology considerations that are not yet fully understood and require additional study and experimentation. The following is a representative list of issues (and management and research challenges) identified in the literature and by the authors. More will evolve as VCE systems become more widely used and more experience is gained in the use of VCEs by IPTs. However, until then additional investigation, understanding, and resolution will be required before VCEs become a widespread reality Individuals VR-induced symptoms and effects (VRISE). Empirical testing confirms that VR systems induce physical symptoms and effects in VCEs. For many VCE participants, symptoms are real and temporary but for others they are severe enough to preclude future involvement. As a result, CAVE VRISE indicators and consequences need to be better understood. (E/PM/ (IT) (I/OB) Visual/body language communications. The extent and variety of visual and body language communication possibilities within a VCE are greatly reduced from more traditional settings in single (local) VCEs where stereo glasses partially cover and encircle participant eyes and the room is darkened. Between local and distributed VCEs, normal communication methods, e.g., hand and arm pointing and gestures, facial expressions, and body position of individuals are totally obscured. Methods need to be found to improve interpersonal communications for IPTs working in a VCE. (E/PM) (I/OB) (CSCW) Teams Small group behavior. According to Slater, et al [27] there has only been limited study of small group behavior and interaction in VCEs when compared to more traditional settings. Further research is also needed to assessing social discomfort levels generated in a VCE caused by participants working concurrently with real people and their avatars (i.e., virtual human representations inserted into VCE displays). (E/PM) (I/OB) (CSCW) Collaboration between remote participants. Reported results strongly suggest that shared and remote VCEs offer substantial benefits. However, more research is needed to find ways to overcome the disadvantages of distance, time, and organizational and natural cultures. (E/PM) (I/OB) (CSCW)

9 Systems Comparison testing of VCE systems. There is a need for a comparative investigation of the advantages and disadvantages of the CAVE to other VCE systems. Other related questions are: (a) Can a single system support all phases of the product development and acquisition process? or (b) Would a combined technology approach be required? (E/PM) (IT) (CSCW) Earlier work by the authors [29] could be used as an empirical test example. Cost benefit analysis. A positive cost/benefit justification may be needed for IPT/management technology acceptance. Basic questions that need to be answered are: (a) What types of work tasks and activities are best suited to VR environment team activities? and (b) Are expensive VCE systems more cost effective when compared to traditional paper drawing or CAD-supported design reviews? (E/PM) (IT) (I/OB) Integrating product and process development. Looking ahead to the future, there is a need to integrate all IPT activities into the total DLC. For another related application, Harrison, et al [3] have developed, in conjunction with Sandia National Laboratories, a virtual collaborative engineering capability that allow distributed, real-time visualization and evaluation of design concepts, manufacturing processes, the total factory, and enterprises in one seamless simulation environment. However, more needs to be done to validate such an integrated system into operational VCE settings. (IT) (CSCW) 7. Summary This paper has pointed out that public and private sector organizations are constantly seeking better methods for improving productivity and effectiveness in task accomplishment, and that several are beginning to use VCE technologies as an enabling technology for IPT activities. Presented was a conceptual model of IPT/VCE interfaces and interactions. Discussed in the context of this model were considerations for integrated and collaborative system and product development that considers the interrelationships between people, work, and technology within the broader context of organizational, social, and technical environments. Briefly described were several VCE technology systems namely the CAVE and the WorkWall, and the results of literature research and case studies of their use by several public and private sector technology-based organizations for project development. Organizations and applications studied included: the Army (future combat and support vehicles), Boeing (commercial aircraft), and DaimlerChrysler and General Motors (automobiles). From these cases situations, and using the VCE conceptual model as a classification taxonomy, operational findings (success factors) and research issues were identified and discussed. Also summarized in the paper were the results of an empirical study conducted by the authors that identified the advantages and disadvantages of the CAVE as a VCE. Unique comparison testing (traditional versus VCE) with operational Army user subjects partially validated the advantages of using this immersive VCE for a conceptual design review of a new weapon system. While a statistical significant improvement could not be proven, users overwhelmingly preferred the CAVE--when compared to a traditional design review method. A study conclusion reached, based primarily on multiple qualitative responses by the IPT user community, was that VCEs offer technology-focused organizations and IPTs a greatly expanded and improved capability for creating realistic simulated model representations. Importantly, VCEs offer IPTs the potential of reducing months and perhaps years from the DLC--with resultant significant cost savings. These virtual models and their displays have resulted in more efficient and effective IPT collaborative environments when compared to more traditional methods (i.e., 2D drawing and computer monitor images). The feeling of immersion and presence by IPTs within VCEs, clearly contributes to users feeling involved in both review and decision processes. Case studies and empirical testing results provide important indicators and support for a positive response to the central question of this research. While only limited quantitative supporting data presently exists, quantitative data in the form of enthusiastic responses of operational program managers, designers, users, and other functional IPT members leaves little doubt as to the value of VCEs. Based on these still preliminary results, the answer is Yes! to the research question: Do 3D VCEs provide an improved environment for IPT team interaction? 8. References [1] A Roadmap for Simulation Based Acquisition: Report of the Joint Simulation Based Acquisition Task Force, DOD, Washington, DC, September 1, [2] M. R. Truelove and B. J. Donlin, SMART: A Historical Perspective, Army AL&T, May-June pp [3] J. P. Harrison, B. Christensen, J. Bianco, and M. Gulli, Virtual Collaborative Simulation Environment For Integrated Product And Process Development, Proceedings of the 5th IEEE International Symposium on High Performance Distributed Computing, 1996, pp

10 [4] P. M. Will, Simulation and Modeling in Early Concept Design: An Industrial Perspective, Research in Engineering Design, 1991, vol. 3, pp [5] S. P. Keller, Simulation-based Acquisition: Real- World Examples, Army RD&A, September-October 1998, pp [6] V. D. Lehner and T.A. Defanti, Distributed Virtual Reality: Supporting Remote Collaboration in Vehicle Design, Computer Graphics and Applications, vol. 17, 1997, pp [7] R. C. Smith, Shared Vision, Communications of the ACM, Special Issue, Collaborative Virtual Design Environments, J. M. Ragusa and G. M. Bochenek, eds., December 2001, vol. 44, no.12, pp [8] R. C. Smith, R. R. Pawlicki, J. Leigh, and D. A. Brown, Collaborative Visualeyes, General Motors Research & Development Center, 2000, pp.1-8. [9] W. M. Newman and M. E. Lamming, Interactive System Design. Addison-Wesley, New York, [10] J. Preece, Human Computer Interaction. Addison- Wesley, New York, [11] K. T. Ulrich and S. D. Eppinger, Product Design and Development, McGraw Hill, New York, [12] A. B. Garcia, R. P. Gocke Jr., and N. P. Johnson Jr., Virtual Prototyping Concept to Production, Fort Belvoir, VA: Defense System Management College Press, [13] L. Shu and W. Flowers, Teledesign: Groupware User Experiments in Three-Dimensional Computer Aided Design, Collaborative Computing, 1994, vol. 1, pp [14] F. P. Sansone and H. M. Singer, Improving Time to Market in Consumer Products, AT&T Technical Journal, May/June, 1992, vol. 71, no. 2, pp [15] J. M. Juran and F. M. Gryna, Quality Planning and Analysis, McGraw-Hill, New York, [16] G. E. Dieter, Engineering Design (3rd Ed), McGraw- Hill, USA, [17] V. Dhar and M. H. Olson, Assumptions Underlying Systems That Support Work Group Collaboration, in Technical Support for Work Group Collaboration, M. H. Olson, Ed., Lawrence Erlbaum Associates, New Jersey, 1989, pp [18] C. Cruz-Neira, D. J. Sandin, and T. A. DeFanti, Surround-Screen Projection-Based Virtual Reality: The Design and Implementation of the CAVE TM, Proceedings of the Computer Graphics International Conference, 1993, pp [19] M. T. Bolas, S. Bryson, and I. McDowell, Virtual Model Displays. Fakespace Systems Inc. Report, [20] J. S. Larson-Mogal, An Immersive Paradigm for Product Design and Ergonomic Analysis, Virtual Reality World, July-August, 1994, pp [21] L. Shu and W. Flowers: Teledesign: Groupware User Experiments in Three-Dimensional Computer Aided Design, Collaborative Computing, 1994, vol. 1, pp [22] J. M. Ragusa, and G. M. Bochenek, special section guest editors, Collaborative Virtual Design Environments (CVDE) Communications of the ACM, December 2001, vol. 44, no. 12, pp [23] G. M. Bochenek and J. M. Ragusa, Study Results: The Use of Virtual Environments for Product Design, IEEE International Conference on Systems, Man, and Cybernetics, 1998, vol. 2, pp [24] P. Banerjee, G. M. Bochenek, and J. M. Ragusa, Analyzing the Relationship of Presence and Immersive Tendencies on the Conceptual Design Review Process, ASME/ACM Journal on Computers and Information Science & Engineering, March 2002, vol. 2, pp [25] G. M. Bochenek-Broecker and K. J. Ciarelli, Using Advanced Collaborative Environments in Developing Army Materiel, Army AL&T, May-June 2001, pp [26] S. Cobb, S. Nichols, A. Ramsey, and J. R. Wilson, Virtual Reality--Induced Symptoms and Effects (VRISE), Presence, April 1999, vol. 8, no. 2, pp [27] M. Slater, A. Sadagic, M. Usoh, and R. Schroeder, Small-Group Behavior, Presence, February 2000, vol. 9, no. 1, pp [28] F. Purschke, M. Schulze, and P. Zimmermann, Virtual Reality-New Methods for Improving and Accelerating the Development Process in Vehicle Styling and Design, Proceedings of the Computer Graphics International Conference, June 22-26, 1998, pp [29] G. M. Bochenek, Comparative Analysis of Virtual 3D Visual Display Systems--Contributions to Cross- Functional Team Collaboration in a Product Design Review Environment, Doctoral Dissertation, University of Central Florida, 1998.