Business Process Transformation - Spring 1996 Business Process Transformation - Spring 1996
Much of the BPR literature consists of describing examples of successful reengineering. As Peter Keen has pointed out, managers love examples. The purpose of the examples is to inspire, to give enthusiasm and courage, and possibly to provide ideas which could be reused. It is not easy to draw general principles from these examples, since the best organisational changes often exploit the unique characteristics of a particular situation. Here are some well known, and some less well known, examples. You will notice that in some of these cases IT is scarcely mentioned, but a little thought will indicate that the changes would not have been possible without supporting IT such as shared data bases or electronic communications.
We studied this example last tutorial. The previous process involved the accounts payable department matching an invoice with a receiving note and a purchase order, and then authorising payment. Much time was spent tracking unmatched documents. The key theme of reengineering was payment on receipt of goods rather than receipt of invoice. IT support was in the form of a shared data base which allowed receiving clerks to match goods with orders as they arrived. Goods were either paid for or sent back immediately. (What effect would this have on the systems of the suppliers?) Staff handling payments were reduced from 500 to 125.
(Source: Hammer chap 2)
IBM Credit finances the hardware, software and services sold by IBM. Approving a request for credit (from the salesperson) took 5 steps (take phone call, check credit rating - on a computer, modify standard contract - another computer, calculate interest charge - a third computer, write quote letter) and 6 days (average). Sales reps were depressed and could lose sales. A tracking system was introduced, which allowed sales reps to check on progress, but slowed things down. investigation showed actual processing took 90 minutes. The process was redesigned to be done by a single person with an improved computer system and access to expert advice. Average turn around time was reduced to 4 hours.
(I once enrolled in a new university course and my first semester subjects in 4 hours - the average time is 4 months. I did this because I knew the process and made sure I didn't let go of any pieces of paper.)
(Source: Hammer chap 2)
Wal-mart sells (among other things) Pampers disposable nappies supplied by Proctor & Gamble. Pampers are not very valuable but take up a lot of space, so inventory management is important. The supply of nappies was reengineered so that Proctor & Gamble now generates orders to itself for shipping nappies to Wal-Mart warehouses, based on an analysis of sale patterns. This drastically reduces inventory and simplifies paper work.
(Source: Hammer chap 3)
Navistar International makes trucks. In this case a process (Goodyear supplying Navistar with tyres) has been shifted from one company to another. Goodyear manages "Navistar's inventory" of tyres and guarantees delivery "just in time". As a result the level of inventory has been reduced from 22 days' supply to 5 days'. Joint overheads are reduced by having one inventory control function rather than two.
(Source: Hammer chap 3)
Danish railways are a vital part of the national infrastructure and the Danish are rightly proud of their design skills. When new 180kph IC (Inter-City) trains were designed for DB (Danish State Railways) they were comfortable, stylish and employed the latest technology. IC trains are heavily used and booking is compulsory. The classic problem with railway booking systems is journeys between intermediate stations. A seat may be booked by one person from Roskilde to Odense, by another person from Horsens to Frederikshavn and be vacant from Odense to Horsens. How do we manage this?
In the early days each station was allocated a number of tickets to sell for onward journeys. Possibly major stations (especially to first station) had a special allocation of tickets for shorter journeys. This meant that many seats were underutilised, and some customers could not purchase seats even though some were available (why?). When telegraph and telephone became commonly available booking could be centralised. Station staff called head office and requested seats. Passengers often had to wait for several hours to find out if a seat were available and manual maintenance of the bookings was a difficult task. Before the journey staff marked those seats which were reserved.
Eventually the problem was almost solved by centralised computer systems with remote terminal access (as operates on NSW Countrylink services). A printed tag is sometimes placed on the seat showing where is booked from to. (In principle this is not necessary if booking is compulsory, because nobody should be looking for a seat, but DB will let you book a "standing place" if the train is full).
The only problem remaining was people who book but don't show up. Airlines solve this problem by overbooking, but this doesn't seem to be acceptable on trains. DB solved the problem with some lateral thinking - they moved the booking computer to the trains. While the train is being prepared for its journey, the bookings are downloaded from the central computer into the train's computer. Displays above each seat show for which part of the journey it is booked. After each stop (these are expresses, they might make 10 stops in a journey) the guard checks tickets of new passengers. The guard has an infra-red remote control and the above seat displays include infra-red receivers. If a booked seat is empty the guard deletes the booking. If there are standing passengers the guard will give the empty seat to one of them and enter a new booking. The booking displays always match the current situation.
A nice system. The only problem was young boys, running through the train and reaching their arms into the toilets, locking the push button doors from the outside. Luckily the novelty of this seems to have worn off, because the only way to regain access to the toilets was to reboot the computer. This deleted all the bookings.
(Source: personal observation)
This is an example of changing a process to support a political decision, or an attempted change of values. When the Whitlam government took power in 1972 it had a policy (as have all previous and subsequent governments) of supporting geographical decentralisation. Given the previous failure of such policies it was decided that some reengineering of the policy implementation process was necessary. The normal process was (and is) that a policy (to fund a new university, say) would be proposed by the relevant department (education) and then sent to the Department of Finance to see if it was properly costed and fitted in with the overall budget. The government decided that a new "super" Department of Urban and Regional Development (DURD) would perform a similar function, checking whether the university or whatever was in the right place - ie whether it would promote the economy of a targeted country region, whether it fitted with a city's public transport plans, etc. DURD was duly set up, given a strong minister (Tom Uren) and staffed with bright young sociologists, town planners and geographers.
This of course caused severe conflict with the powerful Finance and Treasury departments. DURD began to employ more and more economists to help fight treasury and the benefit of a different view was lost. Other departments employed more economists to keep up, and we are still suffering the consequences. (Of course there were other reasons that economists became so popular around that time.) Despite the change of direction and eventual demise of DURD, it was still more successful in promoting regional development (eg in Bathurst and Albury) than most other attempts.
(Source: personal observation)
Oticon is a leading Danish hearing-aid manufacturer. By 1990 they had lost their competitiveness and seemed to have trouble coming up with new products. The hierarchical structure of the organisation was dismantled and people were seen as developing skills portfolios rather than "having jobs". Paper was eliminated and all communication was conducted face to face or electronically. Workers have no fixed work place and carry their few possessions around in a mobile cabinet - a similar system is used in some IBM and consultants' offices, but here the aim is to increase mobility rather than to save space. Projects may be initiated by top management or anyone else. Projects are advertised and people volunteer to work on them. Workers develop their own policy guidelines and in fact change project teams frequently. On the whole the system seems to work.
(Source: Peters chap 13)
Summary of Lecture 3, 16/8/96.
Any questions or comments, just e-mail me.
This page is maintained by
Jim Underwood
who can be reached at
jim@socs.uts.edu.au.
This page was last updated on September 1st, 1996.
http://linus.socs.uts.edu.au/~jim/bpt/lect3.html
