A project-management lesson shows a bar chart, a network and a baseline, then an exam question asks which one you need. The terms can blur together because all three concern time. They are not interchangeable tools, and a colourful schedule does not automatically explain dependencies or preserve the approved plan.
Start with the question you need to answer. When is work planned? What must happen before something else? How uncertain is an estimate? How does the current forecast compare with the approved reference? Those questions give each concept a useful role.
Separate the display from the calculation
A Gantt-style chart displays activities against time. A network representation emphasises relationships between activities or events. PERT is associated with network planning and three-point duration estimates, although software labels and everyday usage can blur these distinctions. PMI's discussion of network conventions explains why the diagram's form and the analysis method should not be treated as the same thing. PMI, PERT Network, CPM Network and Precedence Diagram
When reviewing a chart, ask what information is actually present. Bars alone may show dates without explaining every dependency. A network may show dependencies without providing a convenient calendar overview. Neither automatically proves that the durations are realistic or the resources are available.
Worked example: two parallel activities
Imagine a small fictional learning-resource project. Activity A prepares the outline and takes two days. After A, activity B writes the material in four days, while activity C prepares illustrations in three days. Activity D combines and checks both outputs in two days, so it cannot start until B and C are finished.
Assume finish-to-start dependencies, no waiting gaps, adequate separate resources for B and C, and a simple elapsed-day calendar. The two routes are:
- A → B → D: 2 + 4 + 2 = 8 days.
- A → C → D: 2 + 3 + 2 = 7 days.
The longest route sets an eight-day completion time under these assumptions. Adding all four durations gives eleven days, but that incorrectly treats the parallel work as sequential.
A time chart can display A from day 0 to day 2, B from day 2 to day 6, C from day 2 to day 5, and D from day 6 to day 8. A network makes the reason for D's start clearer: it waits for both predecessors.
A changed estimate can change the controlling path
Now suppose C is re-estimated at five days. The A–C–D route becomes 2 + 5 + 2 = 9 days, longer than A–B–D. The controlling route changes.
This is why a critical-path label is not something to memorise once and leave untouched. It depends on the network and current estimates. The PMI discussion of critical-path calculations explains forward and backward scheduling logic and float. PMI, Critical Path Method Calculations
In a more realistic project, resource constraints and calendars can alter the feasible schedule. The small example deliberately excludes them so you can see the dependency arithmetic. State such assumptions rather than presenting a classroom network as a guaranteed delivery date.
Why the PERT estimate divides by six
The conventional three-point PERT estimate uses optimistic O, most likely M and pessimistic P durations: (O + 4M + P)/6. The denominator is the sum of the weights, 1 + 4 + 1. It is not the number of estimates and not the number of activities. PMI, Five Keys to Estimating
For an original example, use O = 2 days, M = 4 days and P = 8 days. The weighted estimate is (2 + 16 + 8)/6 = 4.33 days, approximately. The ordinary unweighted average would be 14/3 = 4.67 days, so these are different calculations.
The PERT result is an estimate, not a promise. It depends on the inputs and modelling assumptions. Do not attach a completion probability simply because a formula produced a decimal; probability claims require the relevant uncertainty model and analysis.
Keep the baseline separate from the forecast
A schedule baseline is the approved reference against which progress and changes are compared. A current forecast describes where the project is now expected to finish. Updating the forecast does not mean silently rewriting the reference so that every delay disappears. Fleming and Koppelman, maintaining the baseline
In the fictional project, suppose the approved finish is day 8 and the current forecast becomes day 9 after C's estimate changes. Record both. The one-day difference is useful information for review. If a formal change process approves a revised baseline, preserve that decision and its history rather than pretending the original commitment never existed.
For study purposes, make separate columns for baseline, actual progress and forecast. This exposes a common category error: comparing an updated plan with itself and concluding that there is no variance.
Self-test
1. Why is the original example eight days rather than eleven?
Answer: B and C run in parallel after A, and D waits for the later finish. Summing every activity ignores that relationship.
2. What happens when C increases from three to five days?
Answer: A–C–D becomes the longest route at nine days under the stated assumptions.
3. Why divide the PERT weighted sum by six?
Answer: The weights total six. It is a weighted average, not an arbitrary adjustment.
4. Should a late forecast automatically replace the baseline?
Answer: No. The forecast and approved reference have different roles; baseline changes follow the project's change-control process.
Draw the same small project as a network and a time chart, then explain what each makes easier to see. Change one duration and recalculate the controlling path. Our active recall guide can help organise the terminology without confusing the baseline with the latest forecast.
Sources and further reading
Fleming, Q. W., and Koppelman, J. M. (1995). Earned Value: Establish the Project Baseline. PMI.
PM Network (1993). A plea for clarity, 7(10), 41–42. Network conventions.
Kramer, S. W., and Jenkins, J. L. (2006). Understanding the basics of CPM calculations: what is scheduling software really telling you? Project Management Institute. Scheduling calculations.
Stenbeck, J. (2008). Five keys to estimating. Project Management Institute. Three-point estimating.
