A room with a viewer

A room with a viewer

John Audley
Southwestern Bell Telephone Company
500 E. 8* Street, RM 752
Kansas City, Missouri 64106

The Day the Earth Stood Still
Its 1978, and Southwestern Bell is still a part of AT&T. Technology in the engineering department is not quite on the rise at this time, the most sophisticated piece of equipment we own is an electric adding machine. Our engineering drawings are done in pen and ink, or on a typewriter. Our records are drafted in much the same way, on various sizes of media, drawn on a drafting table, distributed to our end users via company mail, on paper. Archived on microfiche, with the originals stored in large tubs that take up nearly as much room as the most sophisticated computers of the day. AM/FM is a way to describe the hi-fi.

Tubs containing maps for 2 wire centers

Its now 1988, and we are an RBOC. Engineering had risen to the technological age, providing one to two dumb terminals in each and every office, dial up of course. Our engineering drawings were now done in pencil, or with a typewriter, on paper. Our records had progressed in some areas to a mechanized plant location records system running on a Data General computer housed in a 2000 square foot room. In other areas, we still were using large format paper drawn on large drafting tables. Distribution in both cases was accomplished via company mail, on paper, and in some instances on microfiche, in the form of aperture cards. Archives were on tape and microfiche, with the originals still stored in the same large tubs as before, taking up nearly as much room as the personnel required to draft the work. AM/FM was a concept to be dreamed of, but still best used to describe the features of a used car.

Now its 1998, on the dawn of a new millennium. Engineers and their clerks are now equipped with PC's on their desks, or UNIX workstations. After years of trying, we now have at our disposal a mechanized drafting tool, using CAD and accessible across a wide area network. Our records system has migrated from paper and pencil, in a variety of formats, to a single CAD format, utilizing these same work stations, archived and stored on disk drives, searchable, easily accessed. Pencils and electric erasers long since discarded, we now modify instead of erase and replace. We the builders of the information super highway now had our work prints and records traveling in the fast lane.

And distribution in 1988 . . . . . . With all of the tools at our disposal, with all the access to the network, and with the sophistication and ease in which data is now passed, with all the technology at our beck and call, the distribution of all of our efforts to our end users, to the people who's jobs depend on accurate records, this distribution is now accomplished . . . . . . . . . .. via company mail, on paper, and in some cases on microfiche, in the form of aperture cards. In 20 years we have indeed come a long way, but unless you were a part of engineering the appearance was that we had not moved an inch.

A Viewer to a Kill
This is where we were. Here, we had a fantastic system for drafting work prints, for updating and maintaining our records system. Surely the distribution of this information would be equally as sophisticated. After all, the records distribution in our company is liken to a company in and of itself. We have a product for which there is a demand from our customers, the outside plant field technician. We have our data in digital format, how can we best provide this service to our customer?

Well, unfortunately our records 'company' was more like a monopoly. Sure we had a product, and indeed there was a demand for this product. But instead of giving our customers what they wanted (and in many cases, needed), we chose to give them what we could with the least amount of effort... .Paper and aperture cards.

Now wait a minute, you say. Isn't that the same output format that you used in a manual environment? Exactly. We took great pains to develop the best AM/FM system we could, and succeeded. And we wound up with a fine system that was easily used, just so long as you had access to our system. And this access must be in the form of a UNIX workstation hooked up to our wide area network. This proved a bit difficult for our customers who were in the field running trouble.

We began looking at several alternatives. We first established that paper and aperture cards had to go away. They are expensive to produce as well as maintain. That given, all solutions pointed to the need for computers in the hands of our users. Since many of our techs are in a somewhat hostile environment, this PC needed to be able to withstand the rigors of the outside world. And it would be beneficial to be able to carry this PC inside a building or manhole and not be tethered to a motor vehicle. Thus our solution was looking at a ruggedized laptop PC.

Now that we had the PC, how best to get the data from the wide area network to the wide-open spaces. Since we are talking about several gigabits of data we looked at two possibilities, intranet and CD. Let's explore these two options.

First, the intranet. Using an intranet to the widespread dissemination of information is definitely in vogue these days in our company. Policies, practices, bulletins, all sorts of information are available, used and encouraged. Could we allow access to this records system via and intranet connection? The answer here is yes, and in fact we are doing that to some extent today. The problem, we found, was not access to the records database but rather access to the wide area network. Many of our locations are not currently connected to the WAN via LAN, and therefore must dial in through a secure Remote Access Server. Sitting at a desk top PC with a 33.6 modem provides access, but when you talk passing graphic images across this modem line, the speed is quite slow. Our field technicians are a different story. In our market area, access to digital cellular technology is not as far reaching as we would like. Therefore, we would be forced to access the RAS via cellular modems. Lets talk about slow. As an experiment, I traveled to the field with my Motorola cellular phone, my laptop, and a modem. After three attempts, I was able to log into the RAS, at a connect rate of 1400. Just to make sure we understand each other, I did not drop a digit in the preparation of this report. That was a connect rate of 1400, not 14,400. After being dropped several times, I was finally able to establish a good connection, access my database, and waited 22 minutes for my first record to appear on the screen. Given today's technology, we could not see the intranet as a viable solution for the outside plant tech.

That took us to CD'S. What we wished to do here was create a CD with some number of records on it. We are still talking about several gig worth of data, and did not wish to carry dozens of CD's around, so we needed the data compressed to be able to fit as many wire centers as possible on a single CD. It needed to be searchable, user friendly, and we needed the ability to red line these drawings without affecting the data. And it had to be able to deal with several different drawing formats, from scanned raster to pure vector to hybrid raster/vector drawings. Most viewing packages allow one format or the other, but not both. Thus we went in search of the ultimate viewing software. This started to remind us of a Don Quitiox quest, until we came across NMT corporation with their FAAR program, which gave us all of the abilities we sought and then some. We could search, tile drawings, print, markup, and distribute to our field forces in a timely and efficient manner.

FAAR Wars
We now had a hardware solution coupled with a software solution. All that was lefl was to deploy these solutions to the field. We then ran into the realization that upper management was right... We were not made of money. Fully equipped with PC, printer, and software, it was going to cost roughly $5,000 per technician to get up and running. Add to that the cost to convert these drawings, and we were looking at a large sum of money. The project could have been scrapped at that time except literally everyone who used records for their jobs that saw the software in action wanted it, begged for it. It every one's mind, this was a clear winner, something that we had needed for a long time. Even our upper management team saw the benefits, but the bottom line kept creeping into our strategy. So our quest became proving this product economically.

Each of our field technicians currently carry a palm top "computer" which is used to input time, pick up jobs, and test our circuits. These units are not nearly as sophisticated as a Pentium based PC and are not capable of displaying graphical information. They also cost about $2,000 per unit. In a field trial conducted in Kansas City, nine technicians were equipped with ruggedized PC's, the FAAR program, and the necessary software for time inputs, testing, and job retrieval. In this 60 day trial, we were able to show that one PC in our technicians hands could accomplish all these tasks, give them access to their records, and unlock a vast potential previously hidden to the field technician. Our distribution plan became one of replacement rather than wholesale buying. Thus, instead of replacing this $2,000 unit with another one, we would replace them with a ruggedized PC. Since the cost difference was $3,000, this became the figure we had to justify.

To prove this additional $3,000 expense, we had to make some pretty rash assumptions. Our first assumption was to say that for a field technician to properly and efficiently do their job, they had to have access to accurate outside plant records. And for these records to be accurate, they would need to be updated at least 4 times per year. Actually, this is what we had been doing for years. Once per quarter, we have been providing our technicians with new records. In some cases, a whole new set, but most times we provided just those records that had been updated since the last revision. These revisions were then filed by the field technician in their truck. According to our maintenance staff, it would take a technician between 2 and 4 hours to file these updates, and the needed to go through a manual process of finding the old sheet or card and then replacing it. If we look at a best-case scenario, this it takes only 2 hours to file these updated records, then it cost our company $150.00 per tech PER QUARTER just to file the most accurate set of records for their use. With 2000 field technicians, that means $1.2 million each and every year JUST to file the records that we sent out. With just that figure alone, we could place 400 laptop PC's in the field in the first year! Updates using CD take somewhat less than one minute.

Aperture Cards for ONE Wire Center

Paper Maps for ONE Wire Center

Now lets look at the cost to output this material. Our pro-rated costs for producing a C size piece of paper is just slightly over $1.00 per sheet. There are approximately 90,000 paper records on our market region, and historical y we have updated approximately 40°/0 of these records each year. This means that to produce one set of updated records for the year, around 36,000 maps must be plotted, for approximately $36,000. Since our average crew size is 15 technicians, we are spending about $540,000 per year just to produce and distribute paper maps to the field. We also have over 25,000 maps on aperture cards. Historically, we produce over 1 million aperture cards per year. Now, the cards themselves are cheap, about 11 cents per card. But we must add in the costs of the machines used to produce these cards, and find that figure to be about $5,000 per month. Thus, the cost for producing these cards adds an additional $170,000 into the mix.

CD Containing Maps for137 wire Centers

Our bottom line is now becoming clear. It will cost our market area nearly $2 million just to produce, distribute, and file our updated maps this year. And next year. And the year after that. In 5 years we will have spent the equivilant of a PC for every field technician in the market area, and we will continue to do this forever. We still have not discussed real time savings for having a system that is searchable, easier to use, or any of the other added benefits that a PC based records system will allow. While we are sure that there will be a tremendous savings in this area it is difficult to put a dollar figure on. But, just say that we will be able to save 15 minutes per tech per week, that's 26,000 hours per year, or $1.95 million dollars per year. Each and every year.

Strict economic analysis reveals that the migration to a new technology will be quite expensive. Those figures are easily proven and easily seen, but where we must concentrate on are those areas which do not present themselves in quite so obvious a manner. Empirical numbers are easy to defend; our challenge comes from assumptions that can be challenged. We do not wish to "what if' this thing to death, but if our assumptions are based on sound, reasonable judgments then they must at least be considered. Professor Carliss Baldwin of the Harvard Business School writes, "Companies that apply strict financial hurdles to new processes will face more competition than those that readily commit to innovate." In the world of telecommunication, competition is key, and the leading edge companies will come out on top. Does this mean that everything new is a benefit, that we must jump on every bandwagon just because it's there? No, but by the same token we must not dismiss something out of hand because on the surface it appears that the status quo is cheaper. From the AM/FM conference in 1997, William Donaldson of US WEST Technologies writes, "Progressive companies are recognizing the need to identi~ strategic, intangible, and corporate benefits that go beyond the scope of a single projector application." He follows up with, "Business managers outside the utility industry find it incredible that the largest utility companies in the world still maintain their inventories of deployed products and services by manually drafiingjust as it was done a hundred years ago."

That is where we really are on this subject. "But we've ALWAYS done it this way" is a common phrase around even the most progressive companies. The fact is, that is normally the best reason anyone can come up with for remaining with the status quo. It's safe, comfortable, and familiar. Proven over time. But, never challenged. In our case, no one who saw and used this system would deny that there would be timesaving for our technicians. The only argument comes from what these expected timesaving would be, which is why I began with a 15-minute savings per week. Fifleen minutes is not much time in the whole scheme of things, but as can be seen, it's enough to at least consider a new way of doing things.

And if we are wrong? If we don't realize a full 15 minute timesaving for every employee? Well, we still have not looked at any of the added benefits beyond a quick records retrieval system. We could talk of the layering capabilities, the ability to turn off all layers save streets and fiber, for example. We could talk of merging different databases into a single system, and not only locating our facilities but also customer information. We could talk of GPS compatibility, a function that does not exist in a paper environment. And we could talk of the additional capabilities that will present themselves in the near future.

For our market region in our company, the approach will be to walk before we run. We have already started with deploying this system to every manager's desk with outside plant cable technicians. It has also been deployed to the maintenance center and to every engineering office. Early in 1998, the system will be given to our cable facility locators, the first real field application, to be used in the vehicles. As for our own technicians, the move will be one of migrating from the hand held units to lap top computers as they need replacement, beginning with cable repair technicians and ending after five years with all outside technicians. Slaying the paper dragon will require an attitudinal change as well as a financial one.

References:

Carliss Baldwin. "How Capital Budgeting Deters Innovation And What To Do About It." Research Technology Management, Nov-Dee, 1991, p. 39-45.
William R. Donaldson. "Why Cash Flow Analysis Stifles Innovation And Unfairly Penalizes AM/FM/GIS." AM/FM International Conference XX, 1996, p. 483-491