CLOSING THE GAP
Page 65
Page 66
If you've noticed an error in this article please click here to report it so we can fix it.
In the last of his series of articles intended to demistify the application of computers to transport operations, Derek Wright advises on how to select fleet management software packages AQUICK look at the available software for the transport operator suggests there are a number of packages that claim to be systems for fleet management. In our experience very few actually offer more qualities than should be found in a vehicle costing package.
Less than a handful of companies offer the ability to assist with load planning, routes and schedules, or attempt to organise performance measures. Even in these few instances these aspects of data processing are usually available as addons, part of a suite of programs, with vehicle costing as a base.
To be fair to the software producers, it has to be a matter of definition — establishing where vehicle costing or transport accounting ends and fleet management begins. The guideline I would apply simply asks: Is the system capable of providing information to help in the company's strategy on distribution and fleet performance?
If it does not then the program output is not strictly management output because the management questions which the system may prompt will invariably require seeking additional information from other sources.
We must recognise that to contemplate employing the benefits of computers to process the company's fleet data and perhaps distribute it to regions or depots represents a major undertaking for any organisation. Whereas the computerisation, say, of fuel monitoring, or maintenance records, is a clear case of improving a present task, many firms are unaware of their true costs. Specifying a package to meet unknown needs can seem to make a difficult task virtually impossible.
Whatever area of business is being considered for possible computer application, a systematic approach is needed and, in a diverse area such as fleet management, essential. It is the only way to ensure that the anticipated savings are possible and the appropriate equipment and operating software is obtained. I propose to run briefly through what might be termed a d-i-y guide to analysing and selecting a System.
The first stage must be to examine in fine detail the way we use information to manage our fleet at the moment, for example: O Establish what paper work exists at present.
• Write out a description of what each section, department or manager is supposed to do (load planning, fleet engineering, order processing, and so on).
• Now go back and write a description of what actually goes on, noting any additional information brought in to assist with each task.
• Work out the costs associated with the paper work of each section, eg staff involved, equipment used.
Once this task has been accomplished, it should be possible to sit down and draw out a plan which lays out the location of each section, manager or other points where decisions about managing the fleet arc made. On this plan arrows can be drawn to show what information passes from one person to another or between sections.
Often at this stage many companies detect where duplication exists and this is an opportunity to think about the paper work as a whole. Usually we find this is the stage where the differences between the "official" data produced by the company and the "unofficial" information managers use in their work start to become apparent. It is essential to recognise that people will still use the unofficial channels of communication even if the new computer-based system produces the same data where the informal way is seen as easier, or even just to maintain social contact with colleagues.
In the course of deciding the "politics" of the company, it is also necessary to come to a conclusion of what the fleet management objectives arc in terms of the aims of the business. Remember that information is not knowledge. Many businesses have computers producing tables of figures on fuel used, drivers' hours records, and so on, which are an inch thick. If the information is never read, it never becomes knowledge. The fleet manager must have to hand what he requires for regular management and be able to check details which arise from that regular data readily and easily.
At this stage we are in a position to say exactly what fleet management means in an organisation. Does it include performance reports of depots by regions, or not? Are the bonus payments for drivers summarised as a league table? Is the driver training giving positive returns in fuel economy? And so on.
We now know exactly what the design of our system should be and this means we can implement a manual system to check it works and establish any teething troubles. Investing a little time here can save time, labour relations problems and expensive modifications to the computer software in the future.
Only after completing this detailed assessment will you know exactly what you think might be computerised. Quite often as a result of the thorough checking simple changes to the existing manual methods can bring major benefits and the need for computerisation can be reduced. The whole activity up to this point can be time consuming and boring, but is essential if a sensible decision is to be made.
Experienced help is available from independent consultants who know the critical fleet management areas and can save time without loss of quality. Working for myself I would recommend an independent analysis. Very often companies find it difficult to trust or have confidence in analysts tied finanically to a computer manufacturer or particular software package!
Armed with this specification the question of equipment and software programs can be answered in the same systematic way. There are usually two opinions' at this stage. The first way is to choose a fleet management package and select a computer that will run it.
The second and opposite view recommends purchasing the computer hardware first and then find a program. written for that machine that will do the task we have defined in our earlier analysis.
Whichever method is chosen, it is unwise to rush out to the nearest software house or computer dealer. You would not buy a truck without deciding payload, capacity and power. Neither should you consider computer based equipment without at least some basic idea of what is required. I do not propose to cover this in fine detail, rather I will try and provide a flavour for the critical areas. Perhaps we can group the major areas for examintion under five headings: capacity, output, cost, communications and software. Capacity: This is the part that usually contains a few surprises. The bottlenecks in the system usually turn out to be getting the data into the system. An experienced computer operator can enter about 5,000 to 6,000 characters an hour, but not for 40 hours every week — you
can rely on around 25 to 30 hours productive data entry time. We can calculate the number of characters from the examination of the paper work described earlier. Remember to allow for the worst case: busy seasons, bank holidays and year ends are typical highload periods for order processing, and the rule of thumb says to add about 15 to 20 per cent as a safety margin.
This will tell us if more than one keyboard/computer is required to enter the quantity of data we have on our fleet. Some, perhaps all, of this data will need to be stored once for use by the computer and again for security in case of machine failure.
In fleet management systems it is unlikely that any other than a Winchester disk would be of sufficient capacity with the security copy being produced on magnetic tape (not the cassettes of home computers). To a degree this explains why most of the products on the market sold as fleet management systems run on minicomputers rather than microcomputers. Output: Once the data is in the system we need to consider how the fleet reports are to be printed or displayed. Often more information comes out than the quantity of data entered — especially on accountancy packages! Here again the actual quantity needs assessing (from our manual version) because while the computer is tied up with printers it is unavailable for the operator to enter new data.
Remember here also to consider the worst case bottleneck. The actual printer performance, or the speed of data communication, if computers are transferring data to regions, will dictate this output tune.
Cost: We can now begin to cost the proposed new fleet management methods and compare this with the costs of the original manual methods. At least a saving of 25 per cent should be aimed
for with SO per cent normally being required.
Communications: Perhaps our reasons for contemplating a computer system were not just ,he speed and accuracy but also the benefit of increased communications which better information technology offers. When a number of disparate functions are grouped, as in fleet management, it may be easier to consider using a system made up of a number of individual microcomputers, say, one at each depot, linked as a network.
A benefit with this approach is that each depot would have the information on its own "fleet" activity. Further, if one machine breaks down another can be used in the interim period of repairs.
Should a number of people need to use the same stored data, a "multi user" computer would be necessary.
Software: The software features decide how the machine behaves in response to users' requests. Unfortunately, much of the software is of poor quality. In the area of fleet management it may be worth getting programs written for your company based on the type of analysis I have already discussed.
Once the list of software requirements is established from the analysis they can be broken down into priority areas such as essential, recommended and nice-tohave. This provides a solid base to enter the computer market place.
Most true fleet management systems are unlikely to cost less than £10,000, but it does not end there. Approaching the computer system market is like a safari — full of trepidation. It is essential to find a company that will provide names of reputable companies using their products. Ask these users about the teething problems. How long before the system was fully operational? Did the operators think the training was adequate? And has the technical back-up been sufficient? Many computers have been bought and installed on a whim. I hope that from the brief description I have given you have an idea of sonic of the factors necessary in a professional approach. Often this stage is missed and many computers remain unused.
We can expect technology to help measure and monitor previously difficult areas as well as help with the bottlenecks through automatic data collection, as discussed in the article on on-board systems. Technology will also change the scope of telecommunications so that with falling equipment costs more computer keyboards and screens will be used in businesses.
Undoubtedly providing information electronically can be an economic way of helping the business, but only if that information is required and I hope this series of articles has helped you put the technology into perspective.
Our own research at the Centre for Transport Studies into the road freight industry and the design and evaluation of other transport systems will continue and, we hope, lead to better quality products being specified.
Remember, however, that computers cannot perform any task we ourselves cannot do. They can do arithmetic quickly and store information and retrieve it reliably, but we are some way from the day when they can think. We are currently looking at the way in which the so-called "expert systems" can be used.
Possible applications are to plan routes for hazardous goods, assist in quoting rates for contracts, help in planning routes for small delivery vans, and calculate truck and car fleet costs.
The available technology is not the problem, obtaining the appropriate equipment for the job is, and it might be necessary to admit your lack of expertise in the area and get outside help — it's likely to be cheaper in the end.