Critical Path Method vs. Critical Chain Project Management
Figure 8 shows an example of a critical path in a traditional project plan. Note that these activities on the critical path are joined by dependencies that connect them end-to-end.
Figure 8: Example of a typical Critical Path plan. Total time = 40d
Figure 9 shows the same critical path, as it would appear in a critical chain plan.
Figure 9: The Critical Path plan shown as a Critical Chain plan. Total time = 30d
CCPM estimated task durations are based on effort only. No allowances are made for interruptions or distractions of any kind. Each task is considered to have a 50% probability of completing on time. Safety, contingency and allowances for delays (stuff happening) are aggregated in buffers at the end of the project. Or, if multiple feeding chains exist, then these allowances are aggregated at the end of each feeding chain. Buffers are sized at 50% of the length of the chains they protect.
Aggressive tasks are often estimated at 50% of the "normal" estimate. Thus the Critical Chain is about half the length of the Critical Path. When a Project Buffer of 50% of the Critical Chain is added, the total committed project length is then about 75% of the same Critical Path committed length as shown in Figure 9.
Note that buffers do not protect individual tasks. Buffers do not belong to management. Buffers can be used by any task and they exist to protect the total project commitment. In other words, the Project Buffer belongs to the customer.
It has been shown that Critical Chain projects, even though shorter, average a 95+% rate of being on time. Many of them complete in half the Critical Path time.