Skip to content
Tutorials Center
My learning

Topics

Lesson 21 of 41 ยท Scheduling and Critical Path

Schedule Compression and Resource Leveling

Learn how to shorten a late schedule with crashing and fast tracking, how to level or smooth resources, and how to choose the cheapest crash using cost slope, with runnable code.

Sooner or later someone says: "We need it two weeks earlier." The schedule you built with CPM shows how long the work takes with the planned resources and logic. To finish sooner you have to change something. This lesson covers the standard options: schedule compression (shortening the critical path) and resource leveling and smoothing (making the schedule feasible with limited people).

Rule one: only the critical path matters

Shortening an activity with float changes nothing. Look first at the critical path and consider the cheapest ways to reduce it. After each change, recompute: another path may now be critical.

Crashing

Crashing adds resources to critical activities to finish them faster, at extra cost: overtime, more people, faster equipment, outsourcing. It works only when the activity can really be shortened by more effort. A task that is mostly waiting (testing a batch that must run overnight) does not speed up with more people.

To choose what to crash, compute the cost slope for each critical activity:

cost slope = (crash cost - normal cost) / (normal duration - crash duration)

It is the extra cost per day saved. Crash the critical activity with the lowest cost slope first, one day at a time, until you reach the target or can crash no more. After each step re-check the critical path.

Python Crash the cheapest critical activity

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The loop handles a key subtlety: after crashing B, the Database path (C) becomes critical too, so later savings must shorten both paths or an activity they share. Watch the critical list change in the output.

Fast tracking

Fast tracking runs activities in parallel that were planned in sequence, such as starting build before design is fully finished. It costs no extra money in resources, but it adds risk and rework: if the design changes, the early build work is wasted. Use it where the dependency is discretionary and risk is acceptable.

Technique Cost Risk Best when
Crashing Higher cost Low to medium Money is available; task is effort-driven
Fast tracking No direct cost Higher (rework) Dependencies are soft
Reduce scope Saves money Value lost Features can be dropped (MoSCoW)
Better estimates / remove waste Low Low Padding or idle waits exist
Python Effect of fast tracking by overlap

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Resource leveling

A schedule can be perfect on paper and impossible in practice if the same person is needed on two tasks at the same time. Resource leveling adjusts start and finish dates to fit resource limits, often delaying the project because it pushes non-critical work later and sometimes critical work too. Resource smoothing is gentler: it uses float only, so the end date never changes, but it cannot remove every over-allocation.

Steps:

  1. Build the resource histogram (see Activities and Resources).
  2. Find periods above capacity.
  3. Move non-critical activities within their float.
  4. If still over capacity, delay lower-priority work, extend duration, or add capacity.
  5. Re-run the schedule and check the critical path again.
Python Level two overlapping tasks for one person

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The lesson: with one developer, the "parallel" Front end and Back end become sequential, adding ten days. Leveling shows the true cost of the shortage; the choice is to accept it, hire, or reduce scope.

The critical chain idea

Critical chain scheduling (Goldratt) accounts for resource limits directly. It removes the safety padding hidden in each activity, pools it into buffers (a project buffer at the end and feeding buffers where paths merge) and tells people not to multitask. Progress is tracked by buffer consumption: green, yellow, red. It addresses student syndrome (starting late because there is time) and Parkinson's law (work expands to fill the time).

Trade-offs to explain to the sponsor

Compression always costs something: money (crashing), risk (fast tracking), scope or quality. A good project manager presents options with costs, not a single yes or no, and lets the sponsor decide what to trade.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Crashing the schedule means...

    Crashing trades cost for time on critical activities.
  2. 2. Which activity should you crash first?

    You get the most time saved per currency unit on the critical path.
  3. 3. What is the main risk of fast tracking?

    Starting dependent work early can force you to redo it.
  4. 4. Resource smoothing differs from leveling because...

    Smoothing works within float; leveling may extend the schedule.

Summary

  • Compress the schedule by crashing (cost) or fast tracking (risk), only on the critical path, recomputing after each step.
  • Cost slope = extra cost per day saved; crash the cheapest first.
  • Leveling fits the schedule to resource limits, often lengthening it; smoothing uses float only.
  • Present trade-offs of time, cost, risk and scope to the sponsor.

Next: Monte Carlo Schedule Simulation