Should you schedule the most important things first? If you have several activities to schedule, it seems like a logical strategy to schedule the higher priority ones before the lesser priority ones, right?
Well that doesn’t always work out. In fact, it is rarely the best strategy for finding the best schedule. If you’re able to maneuver priority, size, best-fit and precedence, you’ll come out on top.
Let me explain:
Suppose you have three activities and each uses the same resource. Let’s say the resource is a tool that is available from noon until 2:00 pm except between 1:00 pm and 1:32 pm when it is being used for something else. The three tasks to be scheduled are shown as follows:
Activity A Duration 40 Minutes Priority 3
Activity B Duration 28 Minutes Priority 2
Activity C Duration 20 Minutes Priority 1
Scheduling these in order of their priority–with the earliest start times means that:
Activity C schedules from noon until 12:20 pm;
Activity B schedules from 12:20 until 12:48 pm; and
Task A does not have any time intervals remaining that are long enough to schedule. Tough luck, Task A!
However, it is easy to see that using a different processing sequence, all three tasks can be scheduled:
Task A schedules from noon to 12:40 pm;
Task C schedules from 12:40 until 1:00 pm; and
Task B schedules from 1:32 until 2:00 pm.
Boom!
In this simple example, using a best-fit approach, rather than a “priority” approach, leads to a better schedule. This becomes even more relevant when you add multiple resources with intermittent usages within the tasks. Scheduling the biggest resource user first is also a logical strategy and in the example, it would also lead to all three tasks being scheduled. However,when there are multiple resources required, the biggest user is also not simple to recognize.
Using only priorities as a criterion for determining a processing sequence is also not a good idea in situations where there are precedence relationships among the activities. Project scheduling and manufacturing have scheduling problems with these types of relationships. If one activity has a high priority but has predecessors with lower priorities, it’s necessary to process the activities in precedence order so that decisions on later activities can be made after those about is predecessors. In this case, a better approach is to order by priority first and then revise that result by inserting predecessors as necessary.
In other words, an intelligent processing sequence when scheduling multiple activities is based on the combination of priority, size, best-fit and precedence—not just priority.