Skip to content
PMMilestoneCalculators
Construction & Project

Schedule crashing vs fast-tracking: buying back time

Compare resource-supported duration reductions with overlapping work, using tested cost slopes and changing critical paths.

By Last updated

Original ink-sketch illustration of schedule crashing vs fast-tracking

Start with the finish that matters

Crashing shortens an activity by buying resources or another credible acceleration measure. Fast-tracking overlaps work that the baseline placed sequentially. Both can recover time, but they change different assumptions. Crashing changes durations and direct costs. Fast-tracking changes the sequence and the exposure to rework. Neither is a synonym for asking everyone to work harder.

Before choosing either, establish a logic-linked remaining schedule. A list of urgent activities does not identify what controls handover. The GAO Schedule Assessment Guide explains why credible dependencies, calendars and critical-path analysis belong together. Its guidance supports the discipline here; all prices and durations below are illustrative project assumptions, not industry benchmarks.

Use the schedule crashing calculator to compare feasible whole-day reductions and their added direct costs. Use the critical path calculator to inspect the network before spending. A cheap day removed from an activity with float may buy no earlier completion at all.

Compare the mechanisms, not just the quoted premium

QuestionCrashingFast-tracking
What changes?Resource-supported durationDependency timing
Example proposalAdd an installation crewRelease fabrication from partial design
Principal checkCan the added resources deliver?Is the overlap technically permitted?
Financial exposureQuoted acceleration premiumCoordination and possible rework
Frequent mistakeShorten a noncontrolling taskRemove a necessary approval boundary

A physical dependency remains physical even when its arrow disappears from a drawing. Concrete curing, safe energization and acceptance inspections cannot simply be deleted to make a schedule fit a deadline. Identify the information required for a partial release and the person authorized to accept its risks.

For resource-driven work, check the remaining quantity and production assumptions in the activity duration calculator. A second crew needs workface access, equipment and supervision. Without those, doubling headcount may add cost while leaving output nearly unchanged. Obtain a delivery plan before adopting the crash duration.

Calculate the cost of a recoverable day

Under a linear endpoint model, cost slope equals (crash cost − normal cost) divided by (normal duration − crash duration). Both costs are complete activity costs, not premiums. The crash duration is the shortest feasible duration, not the number of days removed.

The calculator searches accepted small networks exactly within its whole-day, linear-cost limits. It reschedules each candidate, including parallel critical paths. That promise does not extend to nonlinear shift packages, resource leveling or large unrestricted networks. A subcontractor who sells a complete weekend shift package creates a step cost. An interpolated daily slope may then be the wrong commercial model.

Worked scenario: two branches and shared commissioning

Consider a small equipment installation, measured in working days from boundary zero. Dependencies are finish-to-start with no lag and one calendar applies. The target is to reduce completion from fifteen to eleven days.

ActivityPredecessorsNormal daysCrash daysNormal costCrash cost
A: foundation preparationNone6460008000
B: installationA531000013000
C: electrical preparationNone97900010000
D: commissioningB C4340006000

Enter those four rows and request four recovered days. Cost slopes are $1,000 for A, $1,500 for B, $500 for C and $2,000 for D per activity day saved. Baseline A–B–D takes fifteen days. C–D takes thirteen. Normal direct cost totals $29,000.

C has the cheapest slope, but shortening C alone does not change the controlling fifteen-day path. Shorten A from six to four first. The added cost is $2,000 and both branches now reach commissioning after nine days. With commissioning unchanged, completion becomes day thirteen.

For another project day, shorten B and C together by one day. Their combined premium is $2,000 and both branches become eight days long. Completion becomes day twelve. Repeat the paired reduction to reach eleven: A = 4, B = 3, C = 7 and D = 4. Added direct cost is $6,000, bringing total direct cost to $35,000.

There is an equally priced alternative: A = 4, B = 4, C = 8 and D = 3. This also finishes at eleven days for $35,000. The calculator may return either tied plan. Choose between them using crew availability and commissioning risk rather than claiming that one numerical answer is uniquely correct.

Suppose four earlier working days genuinely avoid $1,800 per day of overhead. Gross savings are $7,200; subtract the $6,000 premium for a modeled net benefit of $1,200. Check that overhead is avoidable. An accounting allocation that remains payable after handover does not become a saving just because a finish date moves.

Worked scenario: overlap a stable design package

A separate project contains ten days of design, eight of fabrication and four of installation. The baseline takes twenty-two working days. Fabrication originally waits for completed design.

The team can release a technically stable package after six design days. Design runs from zero to ten, fabrication from six to fourteen and installation from fourteen to eighteen. The modeled recovery is four days. This overlap requires a start-to-start relationship with a six-day lag, or an explicit partial-release milestone. The finish-to-start calculator cannot directly represent that relationship; model the release separately rather than entering misleading predecessors.

Assume coordination costs $2,000 and avoidable overhead is $1,200 per day. Without rework, the four days save $4,800 and the net benefit is $2,800. Now assume a 25% probability of a design correction costing $6,000 and delaying fabrication by three days. These probabilities are judgments for the illustration, not reported frequencies.

If correction occurs, fabrication ends at seventeen and installation at twenty-one. Only one day is saved. Its $1,200 benefit is outweighed by coordination and rework, producing a $6,800 net loss. Across the two assumed outcomes, expected benefit is 0.75 × $2,800 + 0.25 × negative $6,800 = $400.

The same result follows from expected recovery of 3.25 days, worth $3,900, less $2,000 coordination and $1,500 expected rework. A small positive average is not a guarantee. The sponsor still faces the unfavorable outcome and must judge whether the assumed correction captures the real downside.

Controls worth keeping in a short approval note

  • Freeze the comparison baseline. Record the data date, remaining quantities, calendars and resource assumptions before changing anything.
  • Define release criteria. Fast-tracking needs an approved information package and a stop-work trigger when assumptions change.
  • Price the whole consequence. Include supervision, shift premiums, coordination and rework without counting a cost twice.
  • Protect shared resources. A recovery plan is not feasible if parallel crews need the same crane at the same time.
  • Recheck after implementation. Update actual progress and remaining logic instead of repeatedly showing the original critical path.

A useful approval states the baseline finish, proposed finish, activity changes, added cost, resource commitments and a downside scenario. It should also identify which constraints remain unchanged. That makes the commitment reviewable when conditions change, rather than leaving the team with a date promise and no explanation of how it was supposed to be achieved.

Calculators used in this guide

Frequently asked questions

Can both techniques be combined?

Yes. Analyze the overlap and the added resources separately first, then recalculate their combined effect. Benefits can overlap rather than simply add.

Should the cheapest activity always be crashed?

No. Its reduction must advance the controlling finish. Parallel critical paths may require coordinated reductions across several activities.

Does fast-tracking always cost less?

No. Its immediate premium can be small while coordination and rework exposure exceed the price of controlled acceleration.

Are these working or calendar days?

The scenarios use working-day coordinates on one calendar. Apply the actual calendar before promising a calendar completion date.

Why can the optimizer return a different plan?

Tied minimum-cost plans can exist. Compare their complete durations and costs, then consider operational feasibility rather than assuming a unique allocation.

Sources and standards

About the author

Dr. Hassan Khames Eliwa holds a PhD in Construction & Built Environment and has more than 20 years of international project controls experience. He is Director of Eliwa Project Controls Limited in Auckland and founder of PMMilestone.org and PMMilestone.com.

Buy me a coffee