Here we review the specific recommendations for interdisciplinary research opportunities at the intersection of work activity, information technology, and building technology and physical spaces.
We all perceive that information technology is rapidly changing the nature of work, but there is little quantitative data to support this. An important step would be to build an extensive taxonomy of the kinds of workers, their tasks, their needs for mobility on the job, and their freedom to work at customer sites or at home. An authoritative census, counting the many people who fall into each of these classes, is critical to understanding the trends in the workplace, so as to focus research efforts on where they will have the maximum impact. We are calling for a significant effort that goes well beyond the relatively limited existing telecommuting studies.
Equally important is to begin to create a census of workplace management strategies so as to identify innovative "best of class" leading edge commercial practices and their quantitative evaluations. The collection of such information would enable the development of a case-oriented knowledge base of which techniques have proven to be the most effective. Not only would this help direct the research community, it would provide a very significant resource for educating future generations of information technology developers, building architects, and facilities planners.
It is desirable to capture extensive data sets that describe the physical and environmental conditions inside existing buildings, as a tool both for facilities planning as well as a source of useful case studies. Detailed analyses of those buildings that are known to provide good support for rapid technological changes would be a worthy addition to the body of case study literature emerging from this research community.
The establishment of a coherent research community at the intersection of these technologies would be significantly accelerated through the creation of a shared, moderated repository that collects the known case studies, mobility studies, environmental measurements, user satisfaction studies, and so on. A mixture of quantitative and qualitative evaluation techniques would be developed around these data sets. For example, the existence of unusual patterns in the captured data could be discovered by researchers using quantitative evaluation. These could then be investigated in more detail by interviewing workers in the actual environments in which the measurements were taken.
To scientifically assess the impact of information and building technologies on group and individual work effectiveness, a rigorous process of experimentation will be needed at the intersection of information technology, building technologies, organizational environments, and management culture.
We understand that in order to demonstrate improvement in productivity and effectiveness, you must be able to measure it. A high priority must be the development of a set of metrics for the process of evaluation. This is made especially challenging, because organizational measures of effectiveness may be different from those appropriate for individuals. For example, unplanned, "opportunistic" meetings have been shown to be very important for group effectiveness of knowledge-based workers. Yet this might make individuals appear to be less effective, since they are spending more of their time in unplanned meetings, rather than contributing to traditional task-oriented measures of productivity! More is known today about individual effectiveness than how groups perform together. And those group-oriented studies which have been performed have focused almost entirely on professional level, knowledge-based workers. Little is known about the group effectiveness of those who do more repetitive kinds of work.
Furthermore, we understand that well designed experiments in this domain are especially hard to accomplish because of the complexity of the parameter space and the difficulty of performing controlled experiments.
Three specific kinds of human centered design studies and evaluations are described below:
A goal of much of groupware application development is oriented towards enabling workers to collaborate even when separated in time and in distance. The underlying research question is whether such virtual proximity is an adequate substitute for physical proximity. Do workers perform more or less effectively when they are no longer co-located with their co-workers? Do workers need to work in close proximity for some period of time, to develop friendship, shared goals, and trust with their co-workers, before being able to exploit tools for remote collaboration?
This last question was investigated by Boeing during the design of the 777 aircraft. Design teams worked together in the same location for 18 months before transitioning to a geographically distributed team model. The critical enablers for successful remote collaboration were the close face-to-face interactions in the early stages of the design, coupled to the availability of the entire aircraft design as a computer-based model that could be shared and updated via distributed computer-aided design systems.
Such investigations could be pursued by conducting empirical studies of current work behavior patterns while quantitatively evaluating the benefits of remote work enabled by information technology. An holistic analysis is needed, considering not just the immediate productivity gains or losses, but also the implications that remote work has on transportation planning, environmental impacts, improved quality of life through reduced commute times, and so on. From these studies, researchers could then develop predictive models for the cost effectiveness of new building technologies and alternative physical workspace layouts on worker productivity.
As researchers, we have a generally optimistic viewpoint that information technology and workspace flexibility will improve worker productivity and effectiveness. Yet it is not always the case that more technology is better technology. A good example is the rapid rise of such ailments as carpel tunnel syndrome, which afflicts a growing number of computer users [Tenner 1996]. We must be able to quantify the improvements brought about by new technologies, in information or building technology, and to include health effects in this evaluation.
A strategy for verifying how technology improves worker productivity must be based on defining and validating a broad range of effectiveness measures for people. Having such measures will make it possible to compare alternative points in the technology design space. The measures we are interested in must go well beyond the existing time and motion studies, since simply doing tasks faster is not necessarily correlated with doing them better.
In addition, large-scale user-centered experiments should be designed and executed that embody in scientifically controlled ways new technologies for collaboration, worker mobility, and rapid workspace re-configuration, both inside buildings and out-of-building. Some scenarios for these experiments are explored in more detailed in Recommendations #3 and #4 below.
Finally, novel technologies should continued to be explored, in an effort to identify the critically important workplace applications of the future. These might be new software technologies, such as videoconferencing or tools for moderated group work, or new hardware technologies, such as embedded sensors or higher bandwidth, more robust wireless communications. This short list is by no means meant to be exhaustive.
We believe that human well-being--as indicated by individual effectiveness, comfort, and safety--should be a concern of all technology development. A comfortable and safe workspace will enhance worker productivity. Careful attention must be paid to the physical and mental well-being of workers as new workspace technologies are introduced. The value of such developments should be rigorously demonstrated by careful evaluation of their impact on worker effectiveness. This is a non-trivial undertaking because of the complex interactions that are possible. Working in a telecommuting center might not improve traditional productivity measures of the worker, but the avoidance of stress and the increased time at home brought about by reduced commute times are likely to improve the worker's outlook on life and effectiveness on the job.
To provide effective evaluation, new methodologies for field evaluation of occupant well-being must be developed. For this evaluation to be successful, new worker-centered metrics and new monitoring methods for occupant well-being must be developed. As a counterbalance to the unrelenting introduction of new information technologies and office plan layouts into the workplace, extensive testing will be necessary to assess the possible physical and mental health hazards. Of particular concern to the workspace participants were the health implications of wireless communications and the input/output technologies of future computer systems, such as eye strain induced from displays and airborne toxic substances given off by printers and other plastic coated appliances, the isolation of workers from windows and views, and the mental health implications of smaller work areas with less opportunity for individual control and personalization.
Advances in information and embedded sensor technology will dramatically reduce the cost of in-building controls, greatly enhance their sophistication, and make feasible its large-scale deployment throughout physical building spaces. This will enable more individualized control, leading to improved user satisfaction through better control over visual, acoustic, spatial, and air quality. We must determine how these technologies can be most effectively harnessed to improve the environmental quality and maintainability of physical workspaces.
This can be accomplished by first developing integrated techniques for the measurement of environmental conditions at individual workstations, secondly creating a comprehensive system architectures for collecting occupant feedback on environmental conditions, and finally constructing information systems for informing building management and occupants of the current environmental state (as well as predicting the future state).
Furthermore, advanced electronics and wireless communications will make it possible to conceive of system architectures in which sensors and controls are located on occupants rather than on building walls. This will make it possible to customize the environment to that experienced by each individual within the building.
Much of the existing studies on communication network patterns have focused on the work of traditional computer users, such as software developers. Little is known about the implications of new workloads, such as World Wide Web Access and mobile access to information, on the design of communications networks. A significant research opportunity is to develop new methodologies and evaluation tools to assist in the deployment and management of mobile (and non-mobile) communications network infrastructures inside physical building spaces.
In particular, these developments would include planning and detailed design tools for mobile network infrastructure deployment and management in the three dimensional environment posed by buildings; performance monitoring, evaluation and network diagramming tools; the development of adaptive algorithms and methods to obtain optimal network quality of service at all times, the exploitation of local information in mobile information system access applications, to improve the quality of the data delivered and its responsiveness; and finally, the development of techniques for mobile information access for enhanced collaboration.
We understand that there is a complex interaction between how people work, the tools and technologies they use, and the places in which they use them. An in-depth investigation is needed to understand the interplay and interaction between information technology and the structure of the workspace, to insure effective co-existence.
For example, one could imagine an intelligent building that understood its individual schedules and preferences, and could begin to condition the meeting room before they arrived. It could insure that relevant information could be staged to the computational elements supporting the meeting room. As participants arrive, the room adapts its lighting to the seating of the participants, with separate delivery of task and ambient conditioning. Information appliances they carry with them gain connectivity within the room, and can also control all of its features with the click of key. Non-local participants could also control certain elements of the room, such as a steerable video cameras, if suitably authenticated and with the appropriate access rights. In other words, we seek the development of a reactive and supportive environment and infrastructure when and where it is needed in the building, that can be exploited by individuals and their current collaborators.
Some specific research efforts include the investigation of the impact of building design on wireless communications and distributed building control systems, the development of new design methodologies for wireless-friendly physical spaces, and the evaluation of how wireless communications can best be used to overcome interconnection and access challenges in existing structures. A rigorous process of evaluating the impact of these technologies on workspace flexibility and the effectiveness of work should be pursued.
A related issue is understanding the impact of mobility on worker productivity. Will mobile workers feel more or less isolated? Will they miss the informal interactions of the workspace, or will they spend less time on non-productive activities like gossip and positioning themselves for the corner office? Elimination of a conventional office could lead to less interaction with co-workers, or could lead to more interaction with workers across a wider range once conventional hierarchical work structures are eliminated. The jury is out, but it is clear that intensive investigations are needed to understand the implications for the workforce of the future.
Needed are a concrete set of scenarios that describe ways in which workplaces and work have been transformed by the introduction of new technologies, whether information or building/facilities-based. There should be testable hypothesis about how work will be transformed by these technologies. Some examples of experiments we have considered can be found in Recommendation #4 below.
New technologies should be carefully evaluated for how well they can be deployed in existing building, since the opportunity for creating new buildings is so limited. Similarly, technology developments, like ubiquitous networking and embedded sensors should be evaluated for their impact on emerging building technologies and forms. A related issue is the justification for ever increasing complexity in our systems. While they introduce greater sophistication in the functions they provide, they often lead to systems that are less robust and which may represent safety problems. A bug in a few lines of switching code brought down the AT&T's long distance phone system on the east coast a few years ago. When life and safety issues come into play, the software systems must be developed with a very high level of assurance that they will behave as specified.
Finally, there is some possibility that office buildings in downtown urban cores as we know them today could disappear in the future, as work becomes better able to be performed outside of the traditional office. Nevertheless, the majority of the workshop participants believed that existing structures would be with us for some time to come, and could benefit from the retrofitting of new technologies.
If the ultimate goal of a technology advance is to prove its effectiveness and to be successfully transferred to commercial practice, then a vigorous effort to build, use, and evaluate new information and building technologies in experimental testbeds should be pursued. It is a long tradition in computer science research to develop new technology for use by the research group. This tradition can be extended to university groups working on building system technologies, where the technologies could be deployed in the researchers' university buildings, their homes, satellite offices, etc.
These testbeds would be "showplaces" where the building industry and its clients could see the benefits of the latest technologies and their application. Such testbeds could sustain high rates of technology turnover, much faster than what could be deployed in existing and new commercial buildings. They could form the focus of efforts to educate the building design community, industry managers of information technology, facilities managers, building clients and others as to the practicality and effectiveness of these new technologies. If seeing is believing, then it is only through such testbeds that the executives with the decision making power to equip existing and new buildings with these new technologies will have their minds changed.
Flexible Building Technology Testbeds could be designed to stretch the bounds of reconfigurability, flexibility, and support for mobility to the extreme, in order to explore the possibilities of a totally plugless, totally reconfigurable workspace. These testbeds would exploit wireless technologies for communications and possibly end power distribution, with extensive use of battery operated appliances, and highly repositionable workspace elements such as furniture, lighting, and conditioning. The goal would be to demonstrate an untethered workspace that can be moved and reconfigured at will, touching on the architectural effects of reconfigurability as well as the communications and other technologies needed to support it. Not only will the knowledge workers exploit this infrastructure, but also the building maintenance personnel and building managers.
Testbeds could be selected for different environments, exhibiting different starting assumptions and architectural constraints. For example, a flexible environment for office plans may be quite different than one that explores the options for a space station or a submarine. Also assumptions about the intended occupants would need to be considered in structuring the testbeds: programmers and design engineers have different requirements than clerical workers or those who work in manufacturing environments or those in laboratories or learning spaces.
It should be noted that reconfigurability could also be achieved by moving people to workspaces rather than the other way around, through innovative methods as hoteling or telecommuting centers. An important goal of the testbed should be to enable the effective evaluation of reconfigurable workspaces. This should lead to the development of a solid cost accounting that demonstrates the economic advantage of reconfigurable workspaces, or at the very least, predicts the cross-over points where the new approaches will become economically feasible.
Fully Instrumented/Smart Building Testbeds could focus on pushing flexible environmental control systems to their extremes. This would demonstrate the possibilities for occupant-centered services and sensor-driven controls. Wireless technologies could be used to investigate how such sensor networks could be deployed into existing structures while embedded sensors could be used in purpose built structures for the testbed. Such a testbed would enable extensive sensor data collection and analysis, would provide spatially correlated environmental models of buildings, and could support more effective feedback to and from users. For example, not only could the occupant provide controls to the system, he or she could also see a visualization of the current environmental sensor readings. This would help occupants understand how their settings influence their comfort levels.
Such a testbed could provide an experimental environment in which to develop technologies that allow a building to "think" and adapt to its usage, based on self-analysis and the collection of extensive history of sensor data and other information, and from these, predict future needs. Intelligence within the building would be used to control the building environment to maximize productivity and individual comfort while minimizing the cost and resources to operate the building and its systems.
Linking such data collections could also serve as the basis of experimentation into better building management tools. For example, one of the uses of extensive embedded sensor networks might be to exploit them to predict failure of the building's components or systems. The efficiency of the various systems in the building, such as lighting, electricity, or networking, could also be better analyzed given their extensive instrumentation. Field maintenance personnel could use location dependent data access applications to obtain information about the specific component or section of the building they are currently working on, either by plugging into a data port or through wireless access methods.
There are significant advantages in capturing this data, and making it widely available. Such access would enable other research groups interested in investigating the relationships between workspaces and worker productivity to pursue their research without requiring direct access to the fully instrumented building. By analyzing the data thus collected, it will permit the researchers to propose changes in industrial practice. Furthermore, by collecting such information and the related studies in a common format, and making it widely accessible via means such as the Internet, this could form the basis of a widespread educational effort on the viability of these technologies for improving worker effectiveness.
A common theme within these testbeds would be their evaluation in terms of their support for security, privacy, and worker comfort, safety, and health. The testbeds should also be used to demonstrate that these technology innovations are both economically and environmentally advantageous, and would provide the controlled environment in which to link workspace attributes with occupant performance.
Several of the workshop participants saw the single major impediment to transferring the new technologies for flexible facilities into commercial practice as being the current methods of cost accounting used by the building industry and the decision makers who initiate building projects. Many of the benefits of information technology and reconfigurable building technologies can only be cost justified over the full life of the building, since their key benefit is their ability to ease the evolution of a building's services over time. A critical need is to develop better economic modeling that can convincing show the advantages of considering life cycle costs rather than first costs in choosing technologies for deployment in buildings. Alternatively, tax and accounting policies could be changed, though not by NSF nor the research community, to better incentivize building owners to consider the long view when commissioning buildings. The costing of a building is a complex task, and the impact of technology innovations on their bottom line costs is challenging to quantify. A more comprehensive view that considers the economic externalities is needed.
This is not to say that flexible building systems cannot also improve first costs, just that the savings must be tabulated and presented convincingly. First cost savings can be realized by carefully designing building systems as a coherent whole, rather than as separate subsystems. For example, mechanical heating systems can be merged with facades, or raised flooring for communications could be combined with HVAC distribution, or sunshading and facade orientation can influence HVAC sizing and configuration.
Many building lighting and networking systems are designed with planned redundancy in the event of expansion. With the proper design for accessibility and ease of reconfiguration, it should be possible to eliminate at least some of this infrastructure redundancy and thereby generate up front savings in the building's cost. A common area of cost savings is in improved energy management inside buildings. Other operational savings can be achieved through reduced maintenance and repair costs over the life of the building by designing systems that are more accessible, easily reconfigured, self-diagnosing, and better able to share load.
We believe that physical space design can improve productivity and worker satisfaction, and thereby significantly reduce costs to the users of the building over the long term. Productivity cost savings can come about through increased worker productivity, improved speed and accuracy of task completion, enhanced worker creativity, and reduced absenteeism. The studies must be done to demonstrate these advantages in a convincing way. In a similar vein, health costs can be reduced for the employees in a well designed building, by avoiding "sick building syndrome," and this must be factored into the life cycle costs as well. An attractive and pleasant work environment can be a key factor in retaining employees. Again, long term studies of known good buildings should be undertaken to create the evidence to support these claims.
Over the life of the building, its interior spaces and major subsystems will undoubtedly undergo change and upgrade, thereby leading to significant waste products. A greater investment in higher performance products and systems could reduce these costs, and should be considered in determining the overall costs of design options. Similarly, there are many costs that are incurred whenever major changes take place inside buildings: the costs of reconfiguring spaces; the costs of accommodating changes in functions, densities, workhours; the costs of accommodating rapid changes in technologies, and the costs of building system overload and failure. Reconfigurable building systems, though somewhat more expensive to deploy initially, can significantly reduce these change-related costs over the building's life. Only by collecting the experiences over a significant number of buildings, coupled to a rigorous economic analysis, will a convincing case be built for the economic advantages of reconfigurable building systems.
Last Updated by Randy H. Katz, 22 January 1997