Calibration Management Information System

Background

In 1987, 27 District Workshop in Warminster, England, was (and most probably still is) the largest District Workshop in the United Kingdom's Ministry of Defence.  I was posted to it as an exchange officer in late 1987; my appointment being 2nd in Charge (2IC) and the Production Manager.  The workshop not only repaired the equipment of all units located in the South West of England, a major client being the local School of Infantry, but also undertook programme repair and overhaul work for the entire Ministry of Defence.  As a consequence, every conceiveable type of equipment was repaired there, ranging from weapons, large and small, to vehicles of all sorts to microscopes and electronic test equipment. 

Calibration of Test and Measuring Equipment.

  A significant function carried out in this workshop was the periodic calibration of all the medical and dental as well as the test and measuring equipment held by the military units located in the South West of England.  To assist with this function, there was a system, called CAMIS.  It was comprised of a small ITL mini-computer with three dumb terminals connected to it and a staff of approximately 2 personnel to administer it.  The CAMIS ITL minicomputer was housed in an air-conditioned room and held a database that contained a list of all of the units served by 27 District Workshop, along with all of the equipment that required calibration, details of previous calibrations performed on each equipment, and dates when each equipment was due to be serviced again. 

CAMIS.

  CAMIS would, on a periodic basis, produce a hardcopy list of the equipment that had to be calibrated.  At 27 District Workshop, there was another computer system called ARROW.  It also ran on an ITL mini-computer, much larger than the one used for calibration management.  The purpose of ARROW was to assist the planning and execution of workshop jobs and to provision the spare parts necessary to effect the repairs.  The staff would manually raise a workshop job for each equipment that had to be calibrated from the list.  As a consequence of creating the job, a "call-in" notice would be sent to the unit holding the equipment to ensure it was taken to 27 District Workshop for calibration.  Once the calibration job had been completed, the details of the job would be manually entered into the Calibration Management System and new date for recalibration entered into the database.

In those days, computer systems were developed by an organisation called "The REME Data Centre".  They intended to replace the Calibration Management System with another ITL computer and had estimated it would cost more than a million pounds and take 15 man years of labour to effect. 

My Background

  At my previous posting, I'd been the Commanding Officer of 1 Base Workshop Battalion, located at Bulimba in Brisbane.  It had a Perkin Elmer 32-20 mini-computer that ran an application called MAWD (standing for Machine Assisted Workshop Documentation).  MAWD was similar to ARROW but did not manage stores because (stupidly) in the Australian Army, repair parts are managed by the Supply organisation.  Spare Parts were managed on another Perkin Elmer minicomputer running an application called SCUBA; standing for Stock Controlled Usage Based Articles.

A few years earlier, I had been in charge of EME Systems Development at Army Headquarters so I had come to 1 Base Workshop Battalion with a lot of ideas of what could be done to significantly reduce the clerical effort associated with running an efficient repair operation.  I have long been the advocate of using small systems in the workplace to assist people in the execution of their duties.  As a part of this, I saw the need to have a convenient means of exchanging data between these small systems and the larger systems that collected data for use by management in their decision making.  When I was in charge of EME Systems Development, I had as my assistant, a young officer, 2Lt Stevan Vujovic, who had previously been an apprentice in the workshop platoon I commanded at the Royal Military College, Duntroon.  It was I who had recommended Stevan for officer training.  Whilst he was an craftsman at Duntroon, Stevan had obtained a degree in Physics and was very good at matters relating to computing.  With Stevan's help and mentoring I conceived and developed two EME systems, EMEMic and EMEDATER.  EMEMic was designed to help people with production planning and general job administration in workshops.  It gathered data for EMEDATER, a mainframe system designed to hold data useful for senior management decision making.  When I became CO of 1 Base Workshop, Stevan visited me and introduced me to a programming language called Turbo Pascal.  He also provided me with a library of routines, developed by Blaise Computing in the US, that allowed IBM PCs to communicate using the RS-232 serial socket.  (To learn more about my efforts at 1 Base Workshop, click here.) 

Suffice to say, that when I arrived at 27 District Workshop, I already had the knowledge and a lot of the code modules to connect PCs to the ITL computer networks, ARROW and CAMIS.  ITL terminals were a little more sophisticated than the Perkin Elmer terminals in that they had a microchip that carried out a number of display functions when they received certain escape code sequences through the RS-232.! I worked these out through a bit of reverse engineering.  I had a box made that allowed me to observe the codes that were being sent to the terminal from the minicomputer and the effect they caused on the terminal's screen.  I could also observe the codes being sent from the terminal to the minicomputer.  Doing this allowed me to create tables of codes where I translated the ITL escape sequences to achieve the same effect on the screen of a PC.

Bob Willcox and his Team

  When I arrived at 27 District Workshop with all of these ideas as to how the local workshop computer systems could be improved, the system administrator, Bob Willcox, at the time regarded me with some scepticism and possibly even suspicion.  British Officers were notoriously non-technical.  I think Bob was worried I would break something and he would be left holding the pieces.  After a while though, as I started to burrow into the problems and show Bob I knew what I was doing, he and his assistant, Doug Bonner became caught up in the excitement of discovery and creation.  Without them, and they were incredibly talented, adept persons, it is unlikely I would have enjoyed the success I did.  As a first project, we introduced a Tray-store ordering system that utilised PSION organizers, Hewlett Packard barcode wands and IBM PCs connected to the ARROW network (without the permission of the REME Data Centre!).

CAMIS - The Calibration Managment Information System

Doug Bonner had a friend, Brendan McKeever, who lived locally and was very good with computers and databases; particularly the FoxPro database, a database that behaved like Ashton Tate's DB4 except it could be compiled and did not require users to buy a runtime licence for every program that was run on a different computer.  Thanks to Brendan's generosity and Doug's native ability, Doug learnt how to program in FoxPro in the space of a few weeks.  I wrote the necessary software to emulate a person raising calibration jobs on the ARROW Computer and to check if the jobs had been completed.  When they were completed, my software would download the job data for automatic input into the FoxPro database.  I also wrote the software necessary to download all of the data out of the CAMIS ITL computer for input into the FoxPro database that Doug had written.  In the space of three months we had the system working perfectly.  This system was so successful, it was adopted throughout the whole of the British Ministry of Defence.  The REME Data Centre, the organisation then responsible for systems development, had estimated the new CAMIS would take 5 years to develop and introduce and had budgeted 15 man years of effort to complete the project.  Not only did my staff and I, part time, complete the project in 3 months, we also produced a far superior result to that which the REME Data Centre had planned.  This initiative saved the British Army well over £3 million Sterling in equipment, facilities and programmer's wages.  It also saved £60,000 Sterling per annum in manning costs from that point onwards.  The system ran for 15 years.

Savings Achieved

All manual scheduling and all manual transcription of data between ARROW and CAMIS was elimintated.  The PC used for CAMIS was a 20MHz 286 with a 40MB hard drive.  It cost £1,000 Sterling.  The ITL minicomputer planned for CAMIS was to cost £100,000 Sterling.  The PC operated in a normal room.  The ITL computer required a dust-free airconditioned room.  There was no longer any need for clerks to run the CAMIS computer.  The workload could be coped with by the existing ARROW Staff.  There were approximately 20 sites that would have had to be equipped with CAMIS. 

This project demonstrated what is possible with a small team of capable people embedded within an organisation developing systems as needs dictate.  There was no specification written before the project started.  Everything was debated and decided over cups of coffee and tea in a staff common room.  What resulted was as a consequence of my vision for harnessing small systems and the incredible talent and diligence of my staff.