WBS, Gantt, CPM, PERT & Tollgates

WBS, Gantt, CPM, PERT & Tollgates

Quick Answer: Break project work into a WBS, schedule it on a Gantt chart, find the critical path (CPM) and, when durations are uncertain, use PERT expected times. Advance only after tollgate reviews prove the phase exit criteria—charter goals, data quality, and risk status—are met.

Why Planning Tools Matter in Define (and Beyond)

ASQ CSSGB BoK II.C.5 expects you to apply project planning tools: work breakdown, schedule views, network logic, and phase reviews. A strong charter without a plan becomes a vague aspiration. Planning answers:

1. What work must be done (WBS)?
2. When and in what sequence (Gantt, network, CPM/PERT)?
3. When may we proceed to the next DMAIC phase (tollgates)?

Green Belts rarely build enterprise PMO schedules, but they must plan Measure data collection, Analyze studies, Improve pilots, and Control handoff with enough rigor that Champions can resource the team and detect schedule risk early.

Work Breakdown Structure (WBS)

A work breakdown structure hierarchically decomposes the project into deliverables and work packages. It answers “what must be produced,” not “who is busy this week.”

Typical levels for a Green Belt DMAIC project:

WBS rules of thumb

• 100% rule: Child elements fully cover the parent—no orphan work outside the tree.
• Deliverable-oriented preferred: “Approved MSA report,” not “meet about MSA.”
• Work packages are estimable and assignable: Someone can own duration, effort, and quality criteria.
• Do not confuse WBS with org chart: Structure is by work product, not by department politics.

Worked example — partial Measure WBS:

2.0 Measure
  2.1 Operational definition of “coding defect”
  2.2 Data collection plan (fields, sample size, period)
  2.3 Historical extract + validation
  2.4 Measurement system check (defect classification agreement)
  2.5 Baseline Y chart and stratification pack
  2.6 Measure tollgate package

If “baseline Y” is missing from the WBS, it will also be missing from the calendar—and the tollgate will fail for good reason.

Gantt Charts

A Gantt chart plots tasks as horizontal bars against a calendar. It shows start/finish dates, duration, overlap, milestones, and often percent complete or assigned owner.

Use Gantt charts to:

• Communicate the project timeline to Champions and process owners
• See resource conflicts (same analyst booked on MSA and DOE the same week)
• Track progress against plan (baseline vs. actual bars)
• Mark milestones (charter signed, pilot start, control plan signed)

Gantt does not by itself prove critical path. Bars can look “busy” while the true longest dependent chain is hidden unless dependencies are modeled. Many teams maintain a simple Gantt in a spreadsheet or project tool and keep a separate network for critical-path thinking on complex phases.

Green Belt practicality: For a 4–6 month part-time project, a phase-level Gantt with weekly work packages is usually enough. Over-detailed day-by-day charts that nobody updates become fiction.

Critical Path Method (CPM)

CPM models activities as a network with finish-to-start (or other) dependencies and deterministic durations. The critical path is the longest path from project start to finish—the sequence with zero total float (slack). Delay any critical activity and the project end date slips one-for-one (absent crashing or resequencing).

Key terms

Forward pass computes earliest times; backward pass computes latest times; float = LS − ES (or LF − EF).

Worked example (simplified days):

Path A–B–C–D–F = 5+4+3+6+2 = 20 days
Path A–E–F = 5+3+2 = 10 days

Critical path is A–B–C–D–F. Compressing E does not shorten the project; compressing D does.

Exam and field insight: Critical path is about logic and duration, not about which task “feels” most important. A politically hot meeting with float is not critical; a quiet data extract on the longest chain is.

PERT (Program Evaluation and Review Technique)

PERT adds uncertainty to activity times using three estimates:

• O = optimistic (best reasonable case)
• M = most likely
• P = pessimistic (worst reasonable case)

Expected time:

[ t_e = \frac{O + 4M + P}{6} ]

Variance (per activity, common form):

[ \sigma^2 = \left(\frac{P - O}{6}\right)^2 ]

Sum variances along the critical path (independence assumption) to estimate project-time uncertainty. CSSGB-level questions often test the te formula and the idea that PERT models uncertainty while classical CPM uses single-point times.

Worked example: Data pull estimates O = 3 days, M = 6, P = 15.

[ t_e = (3 + 4\times6 + 15)/6 = (3+24+15)/6 = 42/6 = 7 \text{ days} ]

Use 7 days in the network, not the optimistic 3. Long right tails (messy IT access, vendor delays) inflate te and variance—exactly why Green Belts should not schedule Measure as if systems always cooperate.

CPM vs. PERT (exam contrast)

In practice many teams blend both: network logic from CPM, uncertain tasks estimated with PERT-style three-points.

Tollgates (Phase Gates)

A tollgate (phase-gate review) is a formal decision point between DMAIC phases. The team presents evidence; Champion (and often Master Black Belt / deployment lead) decides go, conditional go, or no-go / recycle.

Typical tollgate questions

Tollgates are not:

• A PowerPoint festival disconnected from the charter
• Automatic calendar events that always “pass”
• Replacements for daily team management

Green Belt behaviors that pass gates

• Tie every claim to charter metrics and operational definitions
• Surface risks and schedule slips early (with recovery options)
• Bring process owner agreement on data and future controls
• Accept recycle when evidence is weak—cheaper than a failed Improve

Putting the Tools Together

1. Charter defines problem, goal, scope, and metrics.
2. WBS lists the work packages needed to hit phase exits.
3. Network / CPM (and PERT where uncertain) sequences packages and finds the critical path.
4. Gantt communicates calendar dates and tracks progress.
5. Tollgates enforce phase quality before more organizational capital is spent.

Common Pitfalls

• WBS that lists meetings but omits deliverables (MSA report, control plan)
• Gantt with no dependencies—everything “starts Monday”
• Treating every red bar as critical path without float analysis
• Using optimistic durations only (PERT’s reason for existence)
• Skipping tollgates because “we’re agile” while still claiming DMAIC rigor
• Planning Improve software builds before Measure proves the Y and MSA

Planning Toolkit Checklist

• WBS covers all phases through closure and documentation handoff
• Dependencies identified; critical path known for the current phase
• Uncertain tasks use three-point (PERT) thinking, not pure hope
• Gantt matches resource reality (part-time belts and SMEs)
• Tollgate exit criteria written before the phase starts
• Champion calendar holds for gates; risks escalated with options

Master WBS, Gantt, CPM, PERT, and tollgates and you convert Define intent into a controllable DMAIC schedule—exactly what II.C.5 tests and what sponsors fund.