Designing programs with flow charts

After completing this lesson you should be able to:

  1. describe what is meant by a program flow chart
  2. sketch the symbols and constructs used in flow charts
  3. state the guidelines for drawing flow charts
  4. design programs using flow charts
  5. use subprocesses in flow charts
  6. use nested loops in flow charts
  7. use multiway selection in flow charts

There are some exercises for you to do and each exercise has a sample answer:

Exercise 1 - a first flow chart
Exercise 2 - a flow chart with subprocesses
Exercise 3 - an advanced flow chart exercise
Exercise 4 - comparing flow charts and pseudocode

When you have finished the lesson you might like to attempt these questions to assess how much you have learned.


Return to the index
Go to the next lesson
Return to the previous lesson


What is a flow chart?

Step-form and pseudocode program designs are both text-based, the statements are written. Flow charts are a graphical method of designing programs and once the rules are learned are very easy to draw. A well-drawn flow chart is also very easy to read since it basically uses just two symbols, two decision constructs. and two iteration constructs:

there are other symbols but the real work is depicted by the two symbols and the constructs. This shouldn't come as a surprise since in the step-form and pseudocode that is what you have been learning.

The language of flow charts

The major symbols are the DECISION (also known as selection) and the SEQUENCE (or process) symbols. The START and STOP symbols are called the terminals. The SUBPROCESS symbol is a variation on the sequence symbol. There are also connectors drawn between the symbols and you will see these used in the examples below. There is at least one other sequence symbol which is used to represent input/output processes but I think it is unnecessary so I don't use it.

There are some important rules concerning the symbols and these rules apply also to other ways of stating algorithms:

Repeat loop. Note that the repeat loop has the process preceding the decision. This means that a repeat loop will always execute the process part at least once. This is an important point to remember because it may not be what you want to do. For instance assume you are a world power in control of an arsenal of nuclear weapons and have written a program to launch missiles in the event of an attack. Your program contains a loop which launches a missile each time you are struck by an enemy missile, for example:

REPEAT
  LAUNCH MISSILE
UNTIL ENEMY STOPS

Is a repeat loop a good idea in this case? Probably not since, if we assume you are not under attack and you run the program, the repeat loop executes the process at least once and you will probably start the next world war. A while loop would be a safer and more humane choice

While loop. The while loop is basically the reverse of the repeat loop, the decision comes first, followed by the process. The while loop is usually written so that it iterates while the condition is true, the repeat iterates until the condition becomes true. An interesting question is: When should a repeat loop be used rather than a while loop? and vice-versa. The while loop should be used when it is possible that the process or processes which are in the scope of the decision (that is, in the loop) may not need to execute. For example assume you have a designed an air-conditioner controller program and the program turns on the compressor while the ambient temperature is above the desired temperature. A while loop is a good choice here since the ambient temperature may be at the desired level before the compressor part of the program is executed. If a repeat loop was used then the compressor would be turned on but it wouldn't be necessary. That would be wickedly ignorant of green sensitivities. A repeat loop would be a good candidate for the kind of situation in which a program needs to check for an external event at least once. For example: assume you have now written a program for a video cassete recorder and it has a menu for doing things like tuning TV channels, setting the date and time, programming events and so on. When the menu is displayed it is a QUIT option along with all the others, the VCR doesn't know which option will be chosen so it stays in the menu mode, that is repeats it, until QUIT is selected.

The IF ... THEN construct is shown here and is also known as the NULL ELSE, meaning that there is no ELSE part. I have use lines with arrow-heads (connectors) to indicate the flow of sequence. Although this is important in flow charts once you have gained some skill in using them and if you draw them carefully you will find that determining the sequence is straight forward. A typical rule is to use arrow-heads on connectors where flow direction may not be obvious.

The IF ... THEN ... ELSE ... construct has a process at each branch of the decision symbol. The only difference here is that each value of the decision  (TRUE/FALSE) has a process associated with it.

Using flow charts to design programs

With other topics I've explained the use of technique with an example. There is something of a precedent there. This flow chart example uses all the symbols except the subprocess symbol.

The algorithm sums all the even numbers between 1 and 20 inclusive and then displays the sum. It uses a repeat loop and contains a null else within the repeat loop.

The equivalent pseudocode is: 

You can see quite clearly from this example what the price of flow charting is. There is quite a bit of drawing to do in addition to writing the legend in the symbols. The pseudocode is quite simple by comparison so why would you use flow charts?

The major reasons are that the flow chart.

Some rules for flow charts

Well-drawn flow charts are easy to read. What must you do to draw well-drawn flow charts? Here are a few rules:

  1. Every flow chart has a START symbol and a STOP symbol
  2. The flow of sequence is generally from the top of the page to the bottom of the page. This can vary with loops which need to flow back to an entry point.
  3. Use arrow-heads on connectors where flow direction may not be obvious.
  4. There is only one flow chart per page
  5. A page should have a page number and a title
  6. A flow chart on one page should not break and jump to another page
  7. A flow chart should have no more than around 15 symbols (not including START and STOP)

If you study the examples here you should see the rules being applied.

Exercise 1

Now it's time for you to try your hand at designing a program using a flow chart.

Draw a flow chart and trace table for the following problem:

There is a sample answer here but try it yourself first.

Flow charts and subprocesses

There is one last topic to do while we are running hot on flow charts - dealing with subprocesses. Remember that when you studied pseudocode you learned about subprocesses and the benefits of using them.

The subprocess is useful because:

In flow charts subprocesses are also useful in dealing with the flow charting rule that a flow chart should have no more than 15 or so symbols on a page.

Here is an example of the use of subprocesses in flow charts:

This is the main page of the flow chart and it contains two subprocess symbols. Each symbol contains some legend which describes briefly what the subprocess does. Each symbol also contains a page reference which indicates where the subprocess flow chart is.

Note that the flow chart has a title and a page number.

and here is page 12!
The Add even number subprocess appears on its own page as indicated by the main flow chart on page 1.

A subprocess flow chart can contain other subprocesses, there is no limit to how deeply these could be nested.

Exercise 2

With your answer for Exercise 1 modify the flow chart so that it has a main flow chart and shows each of the following as subprocess flow charts:

I have a sample answer here but as with all the others resist the temptation to look at this first.

Using nested loops in flow charts

The nested while loop is shown here. This example is much simplified, it doesn't show any initialisation of either of the loops, the outer loop doesn't do any processing apart from the processing the inner loop, neither loop shows any statements which will lead to the termination of the loops.

Each single step through the outer loop will lead to the complete iteration of the innner loop. Assume that the outre loop counts through 10 steps and the inner loop through 100 steps. The sequence in the inner loop will be executed 10 * 100 times. Nested loops will do a lot of work.

The repeat loop shown here, like the while loop example, is much simplified. It does show two processes, sequence 1 and sequence 2, one process in the outer loop and one process in the innner loop.

Like the while loop the nested repeat loop will see a great deal of work done. If the outer loop does a thousand iterations and the inner loops does a thousand iterations then sequence 2 will be executed 1000 * 1000 times.

Using multiway selection in flow charts

The flow chart form of multiway selection is shown here. You can see it how it shows quite clearly the notion of decisions nested within decisions.

If decision 1 is true then sequence 1 is executed and the multiway selection is finished. If decision 1 is false then decision 2 is tested, if this is true then sequence 2 is done and the multiway selection is finished. If decision 2 is false, you get the picture.

Exercise 3

Here is exercise 3 again from one of the pseudocode lessons. It has been only slightly modified:

There is a sample answer here but attempt the exercise for yourself before looking at the sample.

Exercise 4

Rewrite the nested loops and the multiway selection from above as pseudocode.

There is a sample answer here but attempt the exercise for yourself before looking at the sample.


Summary

In this lesson you studied flow charting, which is the third of the program design techniques on our list, there is just one to go. You should have learned

You should also have learned a little bit more about loops and decisions and gained some more practice at designing programs. You will probably agree that flow charts, although easy to read, are a fairly time-intensive way of program design.

The next lesson introduces Nassi-Schneidermann diagrams, another graphical technique.



Return to the index
Go to the next lesson
Return to the previous lesson


This publication is copyright David Beech and Learning Systems 1997-2002
and may not be reproduced by any means without the written permission of
David Beech.
9 Wyndella Street, Tasmania, Australia


db@codelearn.com