Manual methods – part 1 – time and duration

General

Much project management is now carried out using computer software.
However, it is often useful to understand some manual techniques which underpin the former.
These sections try to give a brief indication of some manual methods.

Time and duration

There may be some confusion between Time and Duration. It needs to be made clear that:

Time

People may often say, ‘What is the time for the task?’ this may be interpreted by some as ‘When does it start?’ or ‘When does it finish?’ or ‘How long will it take to do?’ This can be ambiguous. Time strictly refers to a point in time, for example, 23rd May 2007, or 14:35 hours or noon etc. Tasks are indicated by their start times and their finish times.

Duration

This is the passage of time from one moment in time to another. ‘The duration of the task will be 6 days’.
It’s easy for people to get into the habit of saying, ‘What’s the time for the task?’ make sure you clarify what you mean.

One man day

Normal time measures a day as 24 hours. For the purpose of project management a day becomes a ‘working day’ and is normally given as 8 hours. Any extension of this in a single day would be treated as overtime.
Modern planning software can modify the day to suit your own purpose.

Hence, if a task takes one man 24 hours it is not completed in one day but three.
For project purposes ‘one man day’ is eight hours and ignores tea breaks, lunch breaks, holidays, weekends and nights.

In addition, project days are working days only. So, 5 project days would equate to a normal 7 day week as weekends are ignored.
When a project starts it begins at day ‘0’. Day 15 would be 3 weeks later and day 35 would be 7 weeks later etc.
Work is assumed to be spread out evenly over an 8 hour day for ‘one man day’.

Duration

Duration is the passage of time that a task will take to complete. There has to be a little common sense in the units for the time period.
The major unit used is ‘days’. This simple choice is backed up by the need to estimate how long a task will take. It is rarely practical to give estimates that are accurate to the hour and clearly inappropriate to deal in seconds.

The PERT analysis examines the need for providing reasonable estimations that are accurate.

Just remember that every time you increase accuracy by another order of magnitude it will cost you time to get the information and justify it.
Also, your schedule may only be as reliable as your least accurate estimate.
Estimation of task durations must be justifiable.
These may be supported by data based compiled over many years and particular experiences.

If you underestimate then following dependent tasks may be held up.
On the other hand, if you overestimate you may finish early which can also cause problems if others are not ready.
Constant overestimation tends to inflate budgets when you don’t need them, causing cuts in later years.

Examples

Number 1

Let us assume we wish to dig a hole and it is 15000 cubic feet (cu ft).
If we assume one man can dig at 5000 cu ft per day. Normally this would take 3 man days assuming we can measure the volume of the hole that accurately.

If the hole had dimensions 300 ft x 10 ft x 5 ft this would be correct. However, if we can only measure the dimensions to an accuracy of 5% the range of dimensions could be:

315 ft x 10.5 ft x 5.25 ft = 17364 cu ft

And,

295 ft x 9.5 ft x 4.75 ft = 13312 cu ft.

On this basis the job could take anywhere between 2.7 days to 3.5 days.
Hence, we can only realistically quote the duration as 3 days and not 3.0 or 3.00 days.

There will be plenty of other factors to consider, for example.
The weather and the quality of the worker in terms of productivity.

Number 2

Let us say a worker has to lay wooden flooring to cover 3000 sq ft and he can lay it at a rate of 300 sq ft per day.
The duration of the task would be 3000 / 300 = 10 days.

If the floor was 100 x 30 sq ft and the error was 5% the floor area might be:

105 x 31.5 = 3308 sq ft to 95 x 28.5 = 2708 sq ft.

Hence, the duration range might be, 11.0 days down to 9.0 days.
As the job duration gets longer the potential for variation in the time taken gets larger.
Again 10 days is a reasonable estimate. Other factors to consider may be humidity, the quality of the glue, availability of raw materials etc.

Obviously, if you increase the workers for each job the durations will be reduced pro rata.
There will be an optimum duration that you will not be able to lower in some cases.
For example, if you have too many people laying the floor they will have to wait for drying times and there will be the potential to get in each others way.

In the case of the hole you cannot dig out a lower layer until the top layer has been removed, which slows down progress.

When you start to consider costs the worker will be paid whether he or she is able to work or not.
So weather can be a problem as can overestimating. In the latter case you will have paid for workers who may be sitting around with nothing to do if the job completes earlier than expected.

Number 3

If you bid for a fixed price contract you have to have a very good idea how long a contract is going to last and a good method of estimating the resource needed to complete it.
You will have calculated a daily rate of direct costs for the project given that the project will run for a certain period of time.
Any poor estimates may mean the project goes over the total duration and will cost you money.

If you know that the total cost of part of the project is £50,000 and it costs £1000 per day to do it then you must complete within 50 days or lose money.

Clearly, the use of computer software and spreadsheets makes it a lot easier to examine the effects of changes in costs and durations of activities.