The Applications and Markets breakout session took the task of describing the markets and applications for wearable computers. We identified the barriers to entry in our markets, and offered up suggestions as to how to break through some of the barriers. This document will first describe the markets and then describe the typical users (both industrial and individual users), and then go into a discussion of the concerns, obstacles and recommendations.
The list of participants is as follows:
Tom Blackadar tblackadar@bbn.com (Session Chair), Chris Chung cchung@cs.uoregon.edu, Dave Jaffe jaffe@roses.Stanford.edu http:08/22/96/guide.stanford.edu, Brad Rhodes rhodes@media.mit.edu http:08/22/96/www.media.mit.edu/~rhodes, Bill Reedy reedy.bill@sma1.mcclellan.af.mil, Al Stephan als@spdg.com, Roger Cady rcady@tiac.net, Brian Arthur barthur@intermec.com, Malcolm Bauer Malcolm@cs.cmu.edu, Scott Johnson Scottj@isense.com, Ulrich Neumann uneumann@usc.edu, Dave Carroll VIA (Flexible Computers), Bill Yost CECOM, Greg Bowman Boeing, Damian Jensen Boeing , Acorn Pooley Silicon Graphics
As a primary market requirement we stated that the wearable computer must be able to perform the basic operational tasks with little or no connectivity to the infrastructure. As an industry the wearables marketplace is still three plus years away from a larger horizontal market place. This is due to the uncertainty at this moment as to how people will communicate in the mobile workplace. The vertical industries have just started to figure out how they want to communicate and they are implementing solutions now.
Next I will describe the markets and applications which we feel will be the first adopters of wearables. I will list the markets, and then describe the markets and the applications within the markets briefly. The Primary Markets are:
Manufacturing and construction market segments are currently being addressed by the wearable computing industry. Clearly there are advantages to hands free, quick and accurate data collection, and information's distribution in the manufacturing world, which will reduce the number of operations the workers have to perform. We also envision the worker as an information conduit, providing accurate near real-time information on the status of the production line, the machine health, and employee health. When wearable computers penetrate these markets the workers will become more efficient, and will be able to perform their tasks at the next higher level. This will provide significant cost savings for the employer, reduce training times, and increase productivity, and very good employee reuse.
Warehousing is currently using wearables. The group discussed the applications, and the agreement was that warehousing and inventory control would continue to use wearables and as weables costs came down, and functionality increased, that the warehousing applications would adopt them. Warehousing does not need more functionality than it has now, however proving cost pay-back for new feedback and input devices is the way to get into this market.
Maintenance of complex machinery, of "downtime critical" applications, will be targets for the wearable in the maintenance world. This market will require that all records be able to be brought to the industry, that smart, perhaps genetic algorithms be used to walk the user intelligently through the repair tasks, connectivity to remote experts will be a must. Having short video clips though not necessary, would be useful. The wearable gear must be lightweight, not generate any head, must not get in the way (low profile, embedded wiring.).
The healthcare industry desperately needs low cost wearable computing. This is needed in two areas, personal health monitoring, and feedback to the hospital net, and in physician assist systems. If we are to replace the current systems that hospitals use, we must not only replace the clip board, but provide the graphic clipboard, automated filing system, record retrieval, and patient alarm system. There is infrastructure building happening at this time. It is now time to start addressing what to put at the ends of the infrastructure.
Training as a market was described as enhanced work training, and education. This market will use wearable computers to enhance employee training when the cost benefit curves cross over. This should not be confused with the training that can be provided to maintenance workers, and others who use wearables as part of their job.
Finance/Electronic Commerce, here we are addressing the bullpens of the stock exchanges, the major brokerage firms, and the banking industry. The requirements will include full private secure connectivity, with distributed databases, and multi-modal I/O. Being able to switch in and out personal displays with public displays.
The Sales (remote point of sales) and Service Industry market segment are the insurance salespeople, cable repair people, appliance repair people etc.. The paradigms are shifting now where the service people are now becoming a point of sale presence. Cable service people will use wearable computers to get people to sign off on work orders, and then to get them to sign up for the special of the month. Appliance service people will be able to repair a wider range of appliances, faster more efficiently and then be able to turn the call into an additional point of sale. This industry has been waiting patiently for some time on the infrastructure to support the fielded worker. The industry has already seen some payback with application specific uses.
Delivery, where be it, letter carriers, pizza, laundry, bringing wearable computers into these markets will provide new efficiencies. These markets will need a low cost infrastructure for the smaller business. The communications here will be the major component, and bringing electronic commerce will enable the courier to take remote payments.
Entertainment. The working group didn't discuss this topic very much, but this category was continually thrown out. One can envision that providing a director instant feedback mechanisms to be able to modify, add, look at the insertion of graphics, animation's in real-time could help this individual, providing alternate forms of feedback and input to performers, commentators, and support people would aid the production process, and the game industry is already close to wearables with respect to video games and HMDs, and it is just a matter of time before there will be amusement park rides where one puts on HMDs, direction pointers etc.
Military and paramilitary markets are the tactical operations required by the military, the FBI, law enforcement agencies, the emergency medical technicians, and firefighters. Here combinations of health, situation awareness, and geoposition are all critical to the operations of these individuals. Knowing what the other members of the squad are seeing, or how they are performing is critical. Having hands free operation is a must in these applications. The hands are needed for doing the job.
The USERS
Industrial user
In this section we will go into how industry users will use the wearable computers. This is important, (and also incomplete), in determining the requirements of the different market segments. This section will discuss the communication techniques, the demands for data, and the multi-modal requirements of the industrial user.
Communications for the industrial user will be as fast, as slow, simplex, duplex, automated communications, and manual communications. In a nutshell--what works cost effectively reliably and gets the job done, is what will happen. There will be demands to transmit speech, real-time video, interactive real-time video. There will be requirements to only send minimal amount of datagrams one way, or two way, but the cost associated with megabit file transfer won't be tolerated in these situations. There is the infinite threat of LEO satellite systems which is compounding the problems that already exist in the mobile computing industry.
The data requirements as with the communication requirements for the industrial user will be widespread. However, for the industrial user, the wearable computer must be able to function stand-alone. Otherwise, when the communication infrastructure is no present, then the machine becomes a rock. We clarified, that not all the functionality had to be there, but the wearable computer must still provide useful functionality without the communications infrastructure. If the wearable user is a data collector, the data collected must be "held" for as long as the predicted communication outages.
The industrial user can benefit from multiple smart devices, from torque wrenches with computer feedback, to health monitoring aids to detect fatigue. These multi-modal input devices will enable more data to be collected, and thus will enable better decisions to be made as to how tasks are performed. There will be voice, pen-based, keypad, modalities for input, and there will be HMD, monitors, hand held displays, vibration alerts, and audio modalities for output. Wearable computing should not limit the number of devices which can be connected into or through the wearable system.
By doing this one will be able to monitor the environment, the maintenance factor of the machinery, and essentially get the machinery to communicate with the infrastructure. Obtaining multi-modal data will provide is accurate resource allocation, improved safety, and higher productivity for industry.
Individual Users
In this section we explore how individuals will use wearable computers. The individual users who will be the early adopters are the mobile executive, the physical fitness enthusiast, children, and the upscale suburban home owner. The individual users will not require the 400mhz pentium processor. Rather the wearable computer for these individual will meet the basic requirements of power, functionality, and cost.
The sports and health monitoring user will require that the devices be small, and that the data be visible on a watch display, audible feedback, and a means of transmitting the data to a fixed machine. The use of paging technology will be an enhancement that will help sell this technology. The use of sensors must be done wirelessly, and must be small low power and "near" disposable.
The mobile executive will require a much larger mobile computer, which may not be worn, but carried, however the peripheral devices will carried in and must be pocket size. The mobile executive will require paging services, internet voice, or cell voice. The executive will be tied into the office infrastructure, which will provide scheduling, the full mobile office suite, full database access. The computer in this instance will also be the photo-copier , the travel secretary, as well as the financial consultant/accountant.
The upscale suburban user will require that the wearable provide multi-modal functionality. The wearable will have to provide information seemlessly. The wearable will be small, but will interact with "his" infrastructure (e.g. house, personal computer, car, exercise equipment, children, security system). The children will use the wearable for social interactions, entertainment, and for education enhancement. These units will also be multi-modal, and will provide new modes of communication between parent and child.
The individual user will require the wearable to perform machine to machine communication, to provide a mode of electronic commerce, and the wearable will have to make daily life easier.
Concerns and Obstacles
In this section we will describe the obstacles which currently prevent rapid penetration of the wearable into the marketplace. First we will discuss the network, followed by the wearable system architectures, the modality of inputs, and finally the channels of distribution.
Connecting the wearables to a network is a key concept. The concern the group had is with the available of network services, the infrastructure, the cost and pricing to be connected. The current situation is somewhat chaotic, and thus experts are putting together the systems. Security must be addressed up front. As the networks become stable, there is the concern that the network bandwidth will then become the next main issue.
The systems architectures will have to change. The form factors will have to be smaller, lighter weight, consume less power, (8 hours on a single charge is minimum). There will have to be new modes of input into the wearable which have standard like interfaces, and the software re-use will have to be at the function level. Sensors and actuators will have to be part of the architectures. Other infrastructures will have to be put in place such that the wearables can become more environmentally aware of their surroundings. The HMD and display technologies will have to continue to make improvements in size and power, and the cost of the units will have to continue to erode. The systems will have to be consumer modular in nature.
Finally we have to look at the channels of distribution. The wearables must make it into the local office supper store, the video store, the large discount department store. Connections to services must be as easy (if not easier) as cell phone connections are today. The wearables must erode in pricing. If this happens, the volumes will increase dramatically.
Recommendations
In this last section we recommend the follow list of recommendations for people to start to ponder, as work with wearables continues:
Wearables must be wearable!