by Barry E. Clark, PMP, PMI-SP
For the novice – and not so novice scheduler, creating a dynamic, logical, and accurate schedule can be a daunting task. Having in your tool-kit and adhering to a ‘good scheduling practices checklist’ is essential in achieving that goal. Project managers demand and expect accurate schedules that gives them confidence in using it as one of their chief management tools. Professional, well-structured schedules just do not happen by accident. Here are some of the governing principles/ standards that can guide you in what to do/ what not do when creating a project schedule.
- Schedule should map to the WBS, which defines the project’s scope
- Tasks should have a verb in the task name
- Task Duration: Assigned duration should be consistent with the work/scope required to complete the task
- Long durations on discrete tasks should be avoided. Generally no longer than 2 months
- Constraints: Constraints fine-tune a logic-driven schedule by establishing date restrictions based on factors such as component delivery, near term resource availability, or contractual obligations.
- Soft constraints such as Start-No-Earlier-Than and Finish-No-Earlier-Than enable the schedule to be logic-driven. If there is a logical reason/ justification for using a constraint, it should be documented
- Use of hard constraints such as Must-Finish-On, Must-Start-On, Start-No-Later-Than, & Finish-No-Later-Than prevents tasks from being dynamically adjusted by their dependencies (predecessors/ successors)/ network, thus preventing the schedule from being logic-driven. Hard constraints should therefore be avoided
- All schedule tasks, except the first and last, should be linked (at least one predecessor and one successor), creating the schedule ‘network’ which shows logical dependencies. Missing predecessors/ successors will skew the networked logic, creating erroneous forecast dates, and an inaccurate critical path (CP). Project managers and SMEs/ technical leads need to verify the accuracy of the network. Relationship types: There are four scheduling relationship types used for linking: Finish-to-Start (FS), Finish-to-Finish (FF), Start-to-Start (SS) and Start-to-Finish (SF)
- Schedules should be structured with predominantly Finish-to-Start (FS) relationships. (95% recommended)
- Start-to-Finish (SF) relationship type is counter-intuitive (“the successor cannot finish until the predecessor starts” – huh?) and should only rarely be used and with detailed justification
- Lags and Leads: A lag is a delay in the start of a successor task; leads are the opposite – they accelerate the start of the successor. Lags and leads should be used sparingly. Leads are illogical and should be avoided. If a ‘lead’ is absolutely necessary, changing the relationship type to SS with a lag is recommended. Use of lags should be documented.
- Summary Tasks: Summary tasks should not be linked and should not have resources assigned to them. Linking summaries causes errors in the schedule network and dates by preventing the start of other summary groups until all sub-tasks are completed. Summaries are just that – they are intended to only provide a calculated rollup of the subtasks and linking summaries skews the summary % complete calculation.
- Level of effort (LOE) tasks that continue for the life of the schedule should have durations less than the overall duration of the total program to ensure they do not fall on the critical path (CP).
- Schedule structure: Horizontal Integration – shows work is planned in a logical sequence considering interdependencies; work and planning packages. It ensures the overall schedule is logical and depicts schedule dependencies, etc. within the same scheduling level.
- Schedule Structure: Vertical Integration – Shows consistency between various scheduling levels and consistency between various WBS elements in the schedule.
There is a wealth of information on scheduling practices, techniques, standards, etc. that can be quickly found, including PMI Community of Practice, Microsoft Project Users Group (MPUG), LinkedIn scheduling/PM groups, Project Management blogs, webinars, seminars, and the like. Quick internet searches should get you 95% or more of what you’re looking for when developing your schedule. Remember: a good schedule needs to be accurate in scope definition, dynamic in structure, and realistic.