Crashing vs fast-tracking: how to shorten a project schedule

The two schedule compression techniques

Crashing means adding resources to an activity so it takes less time: more people, overtime, an extra shift, faster equipment or paying a supplier for express delivery. The logic of the schedule does not change; durations do. Crashing always costs money, and usually the cost per day saved rises the further you push.

Fast-tracking means doing activities in parallel, or partly in parallel, that were planned one after the other. You change the logic, for example turning a finish-to-start link into a start-to-start link with a lag, or adding a lead. Fast-tracking costs little up front but adds risk: the later activity starts on information that may still change, so you may have to redo work.

Both techniques share one rule: shortening an activity that is not on the critical path does not shorten the project. If you are not sure which activities are critical, the critical path method guide explains how to find them.

Crashing vs fast-tracking at a glance

	Crashing	Fast-tracking
What changes	Activity durations	Links between activities
Main cost	Money for extra resources	Risk of rework
Works best when	Work scales with people or equipment	Activities can safely overlap
Limits	Crash duration; space; people who can be added usefully	How much of the predecessor must be done first
Typical example	Add a second crew to install fittings	Start building while the last design details are finished

Many recovery plans use both: fast-track first where the overlap is low risk, then crash the remaining critical activities in order of cost.

How to compress a schedule step by step

1. Set the target. How many days do you need to save, and by when?
2. Find the critical path and the near-critical paths (those within a few days of it). They will become critical as you compress.
3. For each critical activity, list the options: how many days it can lose (down to its crash duration) and what that costs. Work out the cost slope: cost slope = (crash cost - normal cost) / (normal duration - crash duration).
4. List the fast-tracking options: which sequential pairs can overlap, by how much, and what could go wrong.
5. Buy days one at a time, cheapest first, re-checking the critical path after each change. When two paths are both critical, you must shorten both to gain a day.
6. Stop when you reach the target, when the next day costs more than it is worth, or when the risk is no longer acceptable.
7. Re-baseline only if the change is approved, and record why the dates moved.

Worked example: crashing by cost slope

A project has two parallel paths from start to finish. Path 1 is A (10 days), then B (8 days), then C (6 days): 24 days, the critical path. Path 2 is D (12 days), then E (9 days): 21 days. The client needs the work done in 20 days.

Activity	Normal duration	Days that can be cut	Cost slope (per day)
A	10	3	$500
B	8	2	$300
C	6	2	$800
D	12	2	$400
E	9	2	$250

The cost slope for B comes from its two estimates: $4,000 at 8 days normally, $4,600 at 6 days crashed, so (4,600 - 4,000) / (8 - 6) = $300 per day.

1. Cut B by 2 days ($600). B is the cheapest critical activity. Path 1 is now 22 days, path 2 is 21.
2. Cut A by 1 day ($500). B has nothing left to give, and A is the next cheapest on path 1. Path 1 is 21 days, equal to path 2. Both paths are now critical.
3. Cut A by 1 more day and E by 1 day ($500 + $250 = $750). To gain a day now, both paths must shrink. Path 1 is 20 days (8 + 6 + 6), path 2 is 20 days (12 + 8).

Total extra cost: $600 + $500 + $750 = $1,850 for 4 days. Notice that E, the cheapest activity of all, was useless until path 2 became critical. Cutting it first would have cost money and saved nothing.

Indirect costs change the picture. If site overhead, rented equipment and supervision cost $400 per day, every day saved also saves $400. Cutting B at $300 per day pays for itself. The other days cost more than $400, so the 4 days save $1,600 in overhead against $1,850 in crashing, a net extra cost of $250. That may still be worth it if a late finish triggers penalties or lost revenue.

A fast-tracking alternative. If C could start 3 days before B finishes (an FS link with a 3-day lead), path 1 drops from 24 to 21 days at no direct cost. The price is the chance that late changes in B force rework in C. Combining that overlap with a 1-day crash of E ($250) and a 1-day crash of B ($300) would also reach 20 days.

Common mistakes

• Crashing non-critical work. It feels productive and saves nothing.
• Forgetting the near-critical path. After a few days of compression, a different path drives the finish. Recalculate after every change.
• Assuming people scale linearly. Doubling a crew rarely halves the duration; space, coordination and training get in the way.
• Overlapping work that depends on unstable inputs. Fast-tracking a design that is still changing can cost more time in rework than it saves.
• Ignoring risk. A compressed schedule has less float to absorb surprises. A Monte Carlo run before and after shows how much the confidence in the new date has dropped; the Monte Carlo and P80 guide explains how to read it.

How to do this in Critova

In Critova you can test compression options directly in the activity grid: shorten a duration or change a link to SS with a lag or to FS with a lead, and the critical path, float and finish date update straight away on the Gantt chart. Save a baseline before you start, so you can compare the compressed plan with the original, and run a Monte Carlo schedule risk analysis to see how the P80 date moves. Learn more on the scheduling feature page.