Planning Waves
Rolling wave planning keeps the top of the WBS complete from the first day and details only the next horizon: “the plan is detailed for the immediately upcoming work only and work far in the future is defined at a high level until later in the project life cycle” [1].
1. Detail where you can see
Githens states the core: “A core premise of RWPP is the use of time-phased structure at the high level of the WBS. At the end of each phase, the team develops the work for the next phase” [2].
Visibility comes from completeness at the top: every deliverable is in the tree with a budget, so the 100% rule always holds. Flexibility comes from deferring detail at the bottom. The tree is lopsided by design, deep where the team knows enough and shallow elsewhere.
2. Work packages and planning packages
NASA names the two kinds of leaf: “A WP provides further detail on work content that is considered near-term, while a PP defines far-term work at a summary level” [3].
| Work package | Planning package | |
|---|---|---|
| Horizon | the current wave | a later wave |
| Detail | decomposed into tasks | one summary element |
| Estimate | bottom-up | top-down, from what is known now |
| Next step | executed and tracked | decomposed at the next replanning |
The budget makes a planning package more than a placeholder; without it the estimate has a hole.
3. Running the waves
Githens gives seven steps. Check that the development strategy suits rolling wave; set the time buckets, meaning how many horizons and how long; give every horizon a high-level estimate, and “Do not drill down to details or attempt bottom up estimating”; detail the first wave bottom-up; baseline with the sponsors; execute; then assess what was learned and replan the next horizon. “The work scope baseline (WBS) is modified to reflect the added detail.”
How long is a wave? Githens answers with a boat crossing an unknown sea: with the horizon twenty miles off and the crossing ninety to a hundred and ten, you reset the course at least five times. A wave is as long as you can see ahead.
4. The replanning is a work package
Rolling wave fails predictably: the replanning does not happen. Githens’s defence is to “Establish a work package and fixed date for replanning”, triggered by a date, a deliverable or a percent complete. “The deliverable of the replanning work package is an updated RWPP plan.” It costs several hours to several days. Skipped, it gives way to “code-like-hell” practices.
Each wave changes the WBS, so keep it under configuration management with stable element codes [4]; wave 1’s costs must still roll up in wave 4.
5. Example: Quill 1.0 and 2.0
Quill is an invented word processor for a publishing house. The first wave details its 1.0 branches into work packages. The customer has already asked for tracked changes in Quill 2.0; that stays one planning package with a purpose, an owner, a summary budget and a tentative date, but no tasks. The sponsor sees the whole scope and cost; nobody decomposes work that depends on how 1.0 turns out.
Detail the next wave; hold the rest as a planning package.
Tracked changes is also risky, because a 500-page document must still open within 2 seconds. So the first wave adds one work package, a prototype on a 500-page document, whose deliverable is the knowledge needed to decompose the package at the next replanning. Do not detail the guess; schedule the learning. A Scrum backlog is detailed appropriately in the same way, its large items split just in time [5].
6. When not to use it
Githens lists four assumptions behind linear planning [2]: problems are solved by splitting them, planning and doing are separate, the future is predictable, and projects are simple. On compact, well-defined, modular work they hold: plan up front, because rolling wave only adds replanning overhead. Where they fail, roll: “You can’t schedule a technological breakthrough.” The price is discomfort, as teams manage precise and imprecise data at once.
How solid is this?
- Where it comes from. Githens is a consultant’s conference paper. Its only data is an informal poll of practitioners, and the method is not evaluated. Cite it for what rolling wave is and how to run it, not as evidence that it improves outcomes.
- Where the terms come from. “Planning package” is NASA’s earned-value vocabulary; it does not occur in Githens or in PMI 2001. Rubin never says “rolling wave”; the link to his backlog rule is ours.
- What is illustrative. “At least five times” is the boat analogy’s arithmetic, not a measured rate. Quill is a constructed example.
Acknowledgments
This page adapts material from lectures by Eduardo Miranda and David Root [6] on software project management.
References
- Project Management Institute,Project Management Institute Practice Standard for Work Breakdown Structures, 1st ed. Project Management Institute, 2001.
- G. D. Githens,“Rolling Wave Project Planning,” in Proceedings of the 29th Annual Project Management Institute 1998 Seminars & Symposium, Project Management Institute, Oct. 1998. [Online]. Available: https://www.wcu.edu/pmi/1998/NP02.PDF
- National Aeronautics and Space Administration,“NASA Work Breakdown Structure (WBS) Handbook,” NASA Headquarters, NASA Special Publication NASA/SP-2010-3404, Jan. 2010. [Online]. Available: https://ntrs.nasa.gov/citations/20110012671
- S. Globerson,“Impact of various work-breakdown structures on project conceptualization,” International Journal of Project Management, vol. 12, no. 3, pp. 165–171, Aug. 1994, doi: 10.1016/0263-7863(94)90032-9.
- K. S. Rubin,Essential Scrum: A Practical Guide to the Most Popular Agile Process. in Addison-Wesley Signature Series (Cohn). Addison-Wesley Professional, 2012. [Online]. Available: https://innolution.com/essential-scrum
- D. Root,Managing Software Development. Carnegie Mellon University / Innopolis University lecture series, 2014.
Disclaimer: AI is used for text summarization, polishing and explaining. Authors have verified all facts and claims. In case of an error, feel free to file an issue.