|★ APPLICATIONS ★ BUREAUTIQUE ★ MINICALC|AMSTRAD CPC BOOK THE AMAZING AMSTRAD OMNIBUS) ★|
|Minicalc (Amstrad CPC Book The Amazing Amstrad Omnibus)||Applications Bureautique|
The Minicalc program, which can be very useful for extrapolating trends, allows you a permanent hard copy of its output. You can, however, use it so the results are just shown on the screen. You are given a choice at the start of each RUN. Minicalc offers one of the facilities provided by spread sheet programs.
If you have any stream of data which represents returns of events occurring in sequence, and which appear to indicate a fairly steady development, you'll find applications for this program.
You could, for example, plot the cost of running a car over a two year period and, assuming you kept the same car (and did not do something radical to it like having an accident or replacing the motor), predict with some certainty the running costs for the following year. Car repairs tend to follow a trend which could be characterised by slowly rising costs, partly due to inflation and partly to the increasing age of the car.
Similarly, the number of rejects on a production line, with constantly improving quality control earlier in the production process, should lead to a gradually decreasing rejection rate. Entering known figures for rejects into Minicalc could provide you with an indication of the reject rate for three, six and nine months ahead, assuming your quality control improvement continues. You may well find that such things as the number of person hours lost due to industrial accidents in your plant shows a downward trend. Minicalc is ideal for producing a forward projection of this trend.
Many relationships can be extrapolated with this program, and so long as you do not run the projection too far into the future (watch out for absurd output with 'excess extrapolation'), you should find the information of value.
An example of excess extrapolation would be to enter the growth pattern in passenger use of a privately-owned bus service, until it exceeded the number of people in the area served by the buses, or exceeded the number of people in the area who did not have easy access to alternative means of transport.
To suggest that because your company showed a groww improvement in output of five per cent per month for the last six months that this growth pattern will continue month after month for five years is ludicrous. This would certainly be placing too much relience on a relatively short period of data collection.
Despite these cautionary examples, you'll still find Minicalc a valuable planning tool, especially if you use it to project for time periods which are similar to the time periods over which your entered data has been collected. That is, if you have sales figures from a particular territory for 12 months and you'd like to see how the next 12 months shape up, assuming gross factors remain much the same over the coming year as pertained during the year for which data is available, you could use Minicalc with some confidence. To project the next decade's figures from a single 12 months' return would not be wise.
However, even this long range forecast could be of benefit in highlighting, for example, the residual deterioration in sales from a certain territory. While a one per cent drop per month in sales over a six month period might not seem too critical and could no doubt be blamed on external factors, projecting this for a further five years could highlight the seriousness of the problem.
For example, entering six months' sales figures into the program (assuming the figures were 100 units, 99, 98, 97, 96 and 95) would show an average deterioration of 1.04%. Projecting this trend would show figures of 84 after 12 months, 74 after 24 months and 65 after 36 months — a fall-off of more than a third!
On the other side of the coin, the output of a growing trend can be a very encouraging source of good news. Assuming, for example, you projected future days lost through strike action, after you have followed a year-long process of improving management/worker relations, and entered figures for the last four quarters of 145 hours, 136, 122 and 104 lost, you'd find that if the trend continued over the next four quarters you'd only lose 91 man hours, 80, 71 and 62 respectively. Even if you doubt the reliability of a straight line projection of this type, you will probably agree that at the very least it gives additional information with which to make management decisions and, even if limited, this can be of value.
Although the program listing and output refers to time periods called 'months', it can obviously be altered or taken to refer to any time period you desire — from nanoseconds to years.