Recovering Two Weeks on a Fixed Board Date
Nova Logistics has a 14-week critical path to deliver a warehouse-scanning rollout before peak season. The board has fixed the launch date, and the project has just learned that a hardware supplier is slipping two weeks - which consumes every day of the remaining schedule buffer. The project manager must recover two weeks. Two chains exist. Chain 1, currently critical at 14 weeks: activity A is 5 weeks and can be shortened by up to 2 weeks at $8,000 per week by adding a second developer; activity B is 4 weeks and can be shortened by 1 week for $12,000 total; activity C is 5 weeks and can be shortened by 1 week for $9,000, but the test lead warns that rework probability rises to about 30%. Chain 2 runs 12 weeks; its activity E (7 weeks) is compressible by 1 week for $10,000. The sponsor has authorised $25,000 of additional funding only.
1. Recommend a compression plan and show the arithmetic behind it.
Read the worked answer
Rank the options by cost per week saved:
| Option | Weeks saved | Cost | Cost per week |
|---|---|---|---|
| A (second developer) | up to 2 | $8,000/wk | $8,000 |
| C (with 30% rework risk) | 1 | $9,000 | $9,000 |
| E | 1 | $10,000 | $10,000 |
| B | 1 | $12,000 | $12,000 |
Plan: crash activity A by 2 weeks, cost $16,000.
Check the arithmetic against the whole network, not just the critical path:
- Chain 1: 14 − 2 = 12 weeks
- Chain 2: unchanged at 12 weeks
- Project duration: 12 weeks - the two weeks are recovered, inside the $25,000 envelope, with $9,000 left.
Why not stop at "A is cheapest"? Because compression only buys time if it shortens the project, not just one path. Chain 2 was sitting on two weeks of float; after crashing A, that float is gone and Chain 2 is critical. The residual $9,000 is best held as insurance against Chain 2 rather than spent on activity C, whose 30% rework risk would likely consume far more than one week.
Also confirm the second developer is genuinely available and can be productive in week one. A crash that takes three weeks to ramp up recovers nothing.
2. What new risk does this plan create, and how should the project manager manage it?
Read the worked answer
The plan trades schedule risk for a different set of risks:
- Zero float everywhere. After crashing A, both chains are critical. Any slip on activity D or E now moves the launch date directly. Weekly progress checks on Chain 2 are no longer optional; they become launch-critical monitoring. This is the single most important consequence and the one most often missed.
- Brooks's law / ramp-up. Adding a developer to an in-flight activity adds coordination cost. If onboarding takes a week, effective savings may be closer to one week than two. Mitigate with a defined hand-off and by splitting A on a clean interface boundary rather than by pairing.
- Quality erosion under compression. The same team is being pushed harder on the same work; verify that A's exit criteria are not being met by skipping reviews.
- Supplier risk is unresolved, only absorbed. The two-week slip was absorbed, not eliminated. If the supplier slips again there is no buffer - this belongs in the risk register with an owner and a trigger.
Actions: update the schedule and critical path, re-baseline with sponsor approval, log the new risks with responses, tell the sponsor explicitly that the launch date now has no protection, and spend or reserve the remaining $9,000 deliberately rather than by drift.
3. Before compressing anything, what alternatives should the project manager put in front of the sponsor, and why?
Read the worked answer
Compression is the most expensive lever and the one that adds the most risk, so it should be the last one offered, not the first.
Present, in this order:
- Challenge the date. Does the board know the supplier slipped two weeks? A fixed date set before the slip may not be fixed once the cost of holding it is known. Cheapest option available: a conversation.
- Reduce scope. A phased rollout - high-volume sites in wave 1, the rest after peak - may deliver most of the business benefit with none of the compression cost. Quantify what wave 1 does and does not cover.
- Fast-track instead of crash. Overlapping activities that are currently sequential adds no direct cost, though it adds coordination and rework risk. Only viable where dependencies are soft; check that no mandatory finish-to-start dependency is being violated.
- Expedite the supplier. Paying the hardware vendor to hold the original date may be cheaper per week than crashing the delivery team.
- Crash - the option above.
Then state the recommendation, the cost, and the residual risk in one paragraph.
The judgement being tested: the project manager's job is to surface the trade-off for the person who owns it, not to silently absorb a supplier failure by spending the team's evenings. A crash plan with no alternatives presented looks decisive but removes the sponsor's ability to make the cheapest choice.