Home › Operations Research › Network Analysis › Application of CPM for Time Optimization
Application of CPM for Time Optimization in Construction Projects: A Case Study
A Complete CPM and Time–Cost Trade-Off Analysis with an Illustrative Construction Project
Abstract
Construction projects consist of numerous interdependent activities whose durations and precedence relationships determine the overall project completion time. The Critical Path Method (CPM) is a deterministic project-scheduling technique used to calculate project duration, identify critical activities and determine scheduling flexibility through float analysis. CPM also provides an important foundation for analysing project time–cost trade-offs.
This article explains the application of CPM to construction project scheduling and demonstrates how project duration can be reduced through activity crashing. A clearly identified illustrative construction project is used to demonstrate forward-pass calculations, backward-pass calculations, total float, critical-path identification, cost-slope calculation and crashing decisions.
The numerical example starts with a normal project duration of 28 days and a normal direct cost of ₹6,10,000. Under the specified crashing assumptions, the project can be reduced to 25 days with an additional direct cost of ₹22,000, giving a crashed direct cost of ₹6,32,000. Importantly, the 25-day schedule is not merely a feasible schedule: for the stated 25-day target and the specified linear, integer-day crashing model, the ₹22,000 additional direct cost is the minimum possible among the available crashing alternatives.
The article also distinguishes classical CPM from broader project optimization. CPM determines the schedule for specified activity durations; time–cost optimization additionally requires decisions about alternative activity durations, resource choices, costs and constraints. Established Operations Research literature and peer-reviewed construction research support this distinction.
1. Introduction
Construction projects involve many activities that must be completed in a particular order. Excavation may have to precede foundation work, foundation work may precede structural work, and finishing activities may depend on the completion of structural and installation activities.
Consequently, project duration cannot normally be determined by simply adding the duration of every activity.
If some activities can be performed simultaneously, their durations overlap. If one activity depends on another, however, the successor activity cannot begin until the required predecessor condition has been satisfied.
The Critical Path Method (CPM) provides a systematic network-based approach for analysing these relationships. The foundational work of Kelley and Walker on critical-path planning and scheduling was published in 1959. Their paper described the use of project information and network-based planning to support the coordination and completion of complex engineering projects.
CPM subsequently became an important component of project management and Operations Research. Standard Operations Research textbooks continue to cover project scheduling, PERT/CPM and time–cost trade-off models. For example, Hillier and Lieberman's Introduction to Operations Research includes a network model specifically for optimizing a project's time–cost trade-off, while Taha's Operations Research: An Introduction is a standard Operations Research text emphasizing theory, applications and computation.
In construction management, CPM is particularly useful because project managers must coordinate activities, resources, deadlines and costs. Research in the Journal of Construction Engineering and Management has also treated construction time–cost trade-off as an optimization problem involving resource choices such as crew size, equipment, construction methods and technologies.
2. What is the Critical Path Method?
The Critical Path Method (CPM) is a deterministic project-scheduling technique in which activities, precedence relationships and specified activity durations are represented through a project network.
The main objectives of CPM are to:
- represent project activities and their precedence relationships;
- calculate earliest activity times;
- calculate latest allowable activity times;
- determine activity float;
- identify the critical path; and
- determine the project completion time.
For a given set of activity durations, the critical path is the path through the project network that determines the project completion time.
The critical path is therefore not simply the path containing the largest number of activities. It is the longest-duration path through the network under the specified precedence relationships and durations.
The original CPM literature and subsequent Operations Research textbooks establish this network-based approach to project scheduling.
3. CPM in Construction Project Management
CPM is useful in construction because construction projects normally contain strong precedence relationships.
For example:
Site Preparation → Foundation → Structure → Finishing → Handover
Some activities may also proceed in parallel:
Site Preparation → Foundation → Structure
while simultaneously:
Site Preparation → Procurement/Installation Preparation → Electrical Work
The two branches may eventually merge before finishing begins.
A CPM network makes these relationships visible and allows the project manager to determine which sequence controls completion.
CPM can therefore answer questions such as:
- What is the expected project duration under the specified activity times?
- Which activities are critical?
- Which activities have scheduling flexibility?
- How much float does each activity have?
- Which activities deserve close monitoring?
- Which critical activities might be candidates for acceleration?
The literature also shows that CPM provides a foundation for analysing time–cost trade-offs, although practical construction optimization may require additional models when resources, quality or other constraints are considered.
4. Important Terminology
4.1 Activity
An activity is a task or operation that consumes time and generally requires resources.
Examples include:
- excavation;
- foundation construction;
- structural work;
- electrical installation;
- plumbing;
- painting;
- inspection.
4.2 Predecessor
A predecessor is an activity that must satisfy the required precedence condition before another activity can begin.
4.3 Successor
A successor is an activity that follows another activity according to the specified precedence relationship.
4.4 Path
A path is a sequence of connected activities from the project start to the project finish.
4.5 Critical Path
The critical path is the path that determines the project completion time for the specified network and activity durations.
Under the standard CPM calculation, activities on a critical path have zero total float.
4.6 Float
Float represents scheduling flexibility.
The most commonly used measure here is total float:
\[ TF=LS-ES \]
or:
\[ TF=LF-EF \]
where:
- \(ES\) = Earliest Start
- \(EF\) = Earliest Finish
- \(LS\) = Latest Start
- \(LF\) = Latest Finish
5. Forward Pass
The forward pass calculates the earliest times at which activities can start and finish.
For an activity with duration d:
\[ EF=ES+d \]
For an activity with more than one immediate predecessor:
\[ ES=\max(EF\text{ of all immediate predecessors}) \]
The largest earliest-finish value at the project terminal point gives the project completion time.
6. Backward Pass
The backward pass calculates the latest times at which activities can start and finish without extending the project completion time.
For an activity:
\[ LS=LF-d \]
For an activity having several immediate successors:
\[ LF=\min(LS\text{ of all immediate successors}) \]
The backward-pass results are then used to calculate total float.
7. Total Float and Critical Activities
Total float is calculated as:
\[ TF=LS-ES \]
or:
\[ TF=LF-EF \]
If:
\[ TF=0 \]
the activity is critical under the standard CPM calculation.
A continuous sequence of critical activities forms a critical path.
An important practical point is that an activity with positive float may sometimes become critical later if project conditions change. Therefore, critical-path analysis should be treated as a scheduling analysis that may need to be updated as the project progresses.
8. CPM is Not the Same as Complete Time–Cost Optimization
This distinction is essential.
CPM determines project scheduling for specified activity durations.
Suppose an activity has a duration of 8 days. Classical CPM uses that duration to determine the network schedule.
But a construction manager may ask:
Can the activity be completed in 6 days by using additional resources?
That introduces a second decision:
- what duration should be selected?
- what resource option should be used?
- what additional cost will result?
- does reducing that activity actually reduce the project duration?
This is the time–cost trade-off problem.
The research literature explicitly connects CPM with time–cost analysis while also showing that construction time–cost optimization can require mathematical programming and resource-selection decisions.
Therefore:
\[ \boxed{\text{CPM Scheduling} \neq \text{Complete Time–Cost Optimization}} \]
A more accurate conceptual relationship is:
\[ \boxed{ \text{CPM} \rightarrow \text{Critical Path} \rightarrow \text{Crashing / Time–Cost Analysis} \rightarrow \text{Optimization} } \]
9. What is Activity Crashing?
Crashing means reducing the duration of an activity by selecting an acceleration option, such as additional labour, equipment, overtime, alternative construction methods or other feasible resources.
For the conventional crashing model, each activity may have:
- Normal Time
- Crash Time
- Normal Cost
- Crash Cost
Usually:
\[ Crash\ Time<Normal\ Time \]
and the crash alternative has a higher direct cost.
However, the actual relationship between time and cost in construction need not be perfectly linear. The constant cost-slope model used in introductory CPM analysis is a modelling assumption.
Peer-reviewed construction research has examined more sophisticated time–cost models precisely because real resource-selection problems can be considerably more complex than simple activity-by-activity crashing.
10. Cost Slope
Under the conventional linear time–cost model:
\[ Cost\ Slope= \frac{Crash\ Cost-Normal\ Cost} {Normal\ Time-Crash\ Time} \]
The cost slope represents the additional direct cost per unit of activity-duration reduction under the assumed linear relationship.
For example, if:
- Normal Time = 6 days
- Crash Time = 4 days
- Normal Cost = ₹90,000
- Crash Cost = ₹110,000
then:
\[ Cost\ Slope= \frac{110000-90000}{6-4} \]
\[ =\frac{20000}{2} \]
\[ =\boxed{Rs. 10,000/day} \]
Thus, under the assumed linear model, each one-day reduction costs ₹10,000.
11. Illustrative Construction Case Study
Important Note About the Data
The following case is illustrative.
It is not claimed to be data collected from an actual construction site or copied from a published construction project.
This distinction is important academically. Published construction research can be used to establish the methodology and real-world context, but a numerical example created for teaching should be explicitly identified as illustrative.
This approach is consistent with the distinction between textbook/tutorial calculations and real construction case studies. For example, Khang and Myint applied a time–cost–quality framework to an actual cement-factory construction project and discussed practical difficulties in estimating model parameters from real project data.
12. Activity Data
Consider the following small construction project.
| Activity | Description | Immediate Predecessor | Normal Time (days) | Crash Time (days) | Normal Cost (₹) | Crash Cost (₹) |
|---|---|---|---|---|---|---|
| A | Site preparation | — | 4 | 3 | 40,000 | 46,000 |
| B | Foundation work | A | 6 | 4 | 90,000 | 110,000 |
| C | Structural work | B | 8 | 6 | 160,000 | 200,000 |
| D | Procurement / installation preparation | A | 5 | 3 | 70,000 | 84,000 |
| E | Electrical work | D | 6 | 4 | 80,000 | 100,000 |
| F | Finishing work | C, E | 7 | 5 | 140,000 | 180,000 |
| G | Final inspection and handover | F | 3 | 2 | 30,000 | 36,000 |
13. Normal Project Direct Cost
The normal direct cost is:
\[ 40,000+90,000+160,000+70,000+80,000+140,000+30,000 \]
Therefore:
\[ \boxed{Normal\ Direct\ Cost=Rs. 6,10,000} \]
This is the direct cost represented by the illustrative activity data.
It does not automatically represent the total economic cost of a real construction project because indirect costs, overheads, penalties, financing costs and other project-specific costs are not included in this illustrative dataset.
14. Network Structure
The precedence relationships are:
\[ Start\rightarrow A \]
After A, the project divides into two branches:
\[ A\rightarrow B\rightarrow C \]
and:
\[ A\rightarrow D\rightarrow E \]
Both branches must be complete before F:
\[ C,E\rightarrow F\rightarrow G\rightarrow Finish \]
Therefore, two principal start-to-finish paths are:
Path 1
\[ A-B-C-F-G \]
Path 2
\[ A-D-E-F-G \]
15. Normal Path Durations
Path 1
\[ 4+6+8+7+3 \]
\[ =\boxed{28\text{ days}} \]
Path 2
\[ 4+5+6+7+3 \]
\[ =\boxed{25\text{ days}} \]
The longest path is therefore:
\[ \boxed{A-B-C-F-G} \]
Hence the normal project duration is:
\[ \boxed{28\text{ days}} \]
and the critical path is:
\[ \boxed{A-B-C-F-G} \]
16. Forward-Pass Calculation
Activity A
\[ ES_A=0 \]
\[ EF_A=0+4=4 \]
Activity B
\[ ES_B=4 \]
\[ EF_B=4+6=10 \]
Activity C
\[ ES_C=10 \]
\[ EF_C=10+8=18 \]
Activity D
\[ ES_D=4 \]
\[ EF_D=4+5=9 \]
Activity E
\[ ES_E=9 \]
\[ EF_E=9+6=15 \]
Activity F
F has two immediate predecessors: C and E.
Therefore:
\[ ES_F=\max(18,15)=18 \]
\[ EF_F=18+7=25 \]
Activity G
\[ ES_G=25 \]
\[ EF_G=25+3=28 \]
Therefore:
\[ \boxed{Project\ Duration=28\text{ days}} \]
17. Backward-Pass Calculation
Starting from the project completion time of 28 days:
Activity G
\[ LF_G=28 \]
\[ LS_G=28-3=25 \]
Activity F
\[ LF_F=25 \]
\[ LS_F=25-7=18 \]
Activity C
\[ LF_C=18 \]
\[ LS_C=18-8=10 \]
Activity E
\[ LF_E=18 \]
\[ LS_E=18-6=12 \]
Activity B
\[ LF_B=10 \]
\[ LS_B=10-6=4 \]
Activity D
\[ LF_D=12 \]
\[ LS_D=12-5=7 \]
Activity A
A has two immediate successors, B and D.
Therefore:
\[ LF_A=\min(LS_B,LS_D) \]
\[ LF_A=\min(4,7)=4 \]
Thus:
\[ LS_A=4-4=0 \]
18. CPM Schedule Table
| Activity | ES | EF | LS | LF | Total Float |
|---|---|---|---|---|---|
| A | 0 | 4 | 0 | 4 | 0 |
| B | 4 | 10 | 4 | 10 | 0 |
| C | 10 | 18 | 10 | 18 | 0 |
| D | 4 | 9 | 7 | 12 | 3 |
| E | 9 | 15 | 12 | 18 | 3 |
| F | 18 | 25 | 18 | 25 | 0 |
| G | 25 | 28 | 25 | 28 | 0 |
Therefore:
\[ TF_A=TF_B=TF_C=TF_F=TF_G=0 \]
while:
\[ TF_D=TF_E=3 \]
Hence:
\[ \boxed{Critical\ Path=A-B-C-F-G} \]
and:
\[ \boxed{Normal\ Project\ Duration=28\ days} \]
19. Cost-Slope Calculation for Every Activity
Activity A
\[ Cost\ Slope_A= \frac{46,000-40,000}{4-3} \]
\[ =\boxed{Rs. 6,000/day} \]
Activity B
\[ Cost\ Slope_B= \frac{110,000-90,000}{6-4} \]
\[ =\boxed{Rs. 10,000/day} \]
Activity C
\[ Cost\ Slope_C= \frac{200,000-160,000}{8-6} \]
\[ =\boxed{Rs. 20,000/day} \]
Activity D
\[ Cost\ Slope_D= \frac{84,000-70,000}{5-3} \]
\[ =\boxed{Rs. 7,000/day} \]
Activity E
\[ Cost\ Slope_E= \frac{100,000-80,000}{6-4} \]
\[ =\boxed{Rs. 10,000/day} \]
Activity F
\[ Cost\ Slope_F= \frac{180,000-140,000}{7-5} \]
\[ =\boxed{Rs. 20,000/day} \]
Activity G
\[ Cost\ Slope_G= \frac{36,000-30,000}{3-2} \]
\[ =\boxed{Rs. 6,000/day} \]
The complete cost-slope table is:
| Activity | Normal Time | Crash Time | Maximum Reduction | Cost Slope |
|---|---|---|---|---|
| A | 4 | 3 | 1 day | ₹6,000/day |
| B | 6 | 4 | 2 days | ₹10,000/day |
| C | 8 | 6 | 2 days | ₹20,000/day |
| D | 5 | 3 | 2 days | ₹7,000/day |
| E | 6 | 4 | 2 days | ₹10,000/day |
| F | 7 | 5 | 2 days | ₹20,000/day |
| G | 3 | 2 | 1 day | ₹6,000/day |
20. Which Activities Should Be Crashed?
The current critical path is:
\[ A-B-C-F-G \]
Therefore, if the objective is to reduce the project completion time, the relevant activities must belong to a path controlling project completion.
The critical-path activities and their cost slopes are:
| Critical Activity | Cost Slope |
|---|---|
| A | ₹6,000/day |
| B | ₹10,000/day |
| C | ₹20,000/day |
| F | ₹20,000/day |
| G | ₹6,000/day |
The cheapest critical activities are A and G.
However, crashing must always be followed by a network re-evaluation.
21. First Crashing Step
Crash Activity A:
\[ 4\rightarrow3\text{ days} \]
Additional direct cost:
\[ Rs. 6,000 \]
New path durations:
Path 1
\[ 3+6+8+7+3=27 \]
Path 2
\[ 3+5+6+7+3=24 \]
Therefore:
\[ \boxed{Project\ Duration=27\ days} \]
The critical path remains:
\[ A-B-C-F-G \]
22. Second Crashing Step
Activity A is already at its crash time.
The next lowest-cost activity on the critical path is G:
\[ 3\rightarrow2\text{ days} \]
Additional cost:
\[ Rs. 6,000 \]
New Path 1:
\[ 3+6+8+7+2=26 \]
New Path 2:
\[ 3+5+6+7+2=23 \]
Therefore:
\[ \boxed{Project\ Duration=26\ days} \]
Cumulative additional cost:
\[ Rs. 6,000+Rs. 6,000 \]
\[ =\boxed{Rs. 12,000} \]
23. Third Crashing Step
Activity A and G have reached their crash times.
The next cheapest activity on the critical path is B:
\[ 6\rightarrow5\text{ days} \]
Additional cost:
\[ Rs. 10,000 \]
New Path 1:
\[ 3+5+8+7+2 \]
\[ =\boxed{25\text{ days}} \]
New Path 2:
\[ 3+5+6+7+2 \]
\[ =23\text{ days} \]
Therefore:
\[ \boxed{Project\ Duration=25\ days} \]
Cumulative additional direct cost:
\[ 6,000+6,000+10,000 \]
\[ =\boxed{Rs. 22,000} \]
24. Crashed Direct Cost
Normal direct cost:
\[ Rs. 6,10,000 \]
Additional crashing cost:
\[ Rs. 22,000 \]
Therefore:
\[ Crashed\ Direct\ Cost = 610,000+22,000 \]
\[ =\boxed{Rs. 6,32,000} \]
Thus, the illustrative schedule changes from:
\[ \boxed{28\ days,\ Rs. 6,10,000} \]
to:
\[ \boxed{25\ days,\ Rs. 6,32,000} \]
25. Is the 25-Day Schedule Actually Minimum-Cost for a 25-Day Target?
This question is important.
Simply finding one feasible 25-day schedule does not automatically prove that it is the cheapest 25-day schedule.
We can, however, prove the result for this specific illustrative model.
To reduce the original critical path from 28 days to 25 days, we need:
\[ 28-25=3 \]
days of reduction on the controlling path.
The available maximum reductions and costs on the initial critical path are:
| Activity | Maximum Reduction | Cost per Day |
|---|---|---|
| A | 1 day | ₹6,000 |
| B | 2 days | ₹10,000 |
| C | 2 days | ₹20,000 |
| F | 2 days | ₹20,000 |
| G | 1 day | ₹6,000 |
The three cheapest available one-day reductions are:
- A = ₹6,000
- G = ₹6,000
- B = ₹10,000
Therefore:
\[ 6,000+6,000+10,000 \]
\[ =\boxed{Rs. 22,000} \]
Any alternative that replaces one of these reductions with a C or F reduction would increase the additional direct cost because C and F have a cost slope of ₹20,000 per day.
The alternative of using D or E cannot reduce the controlling path from 28 to 25 days because neither activity belongs to the initial critical path and both have positive float.
Therefore, under the following assumptions:
- integer-day duration reductions;
- linear cost slopes;
- no resource constraints;
- no additional precedence changes;
- no indirect-cost effects;
- no quality or safety constraints;
- no nonlinear crash costs; and
- the stated activity data,
the minimum additional direct crashing cost required to achieve a 25-day project duration is:
\[ \boxed{Rs. 22,000} \]
Thus the 25-day schedule is minimum-direct-cost for the specified 25-day target within this illustrative model.
This is a much stronger and more precise statement than simply calling the schedule “optimized.”
26. Quick Summary of the Case Study
| Parameter | Normal Schedule | 25-Day Crashed Schedule |
|---|---|---|
| Project Duration | 28 days | 25 days |
| Time Saved | — | 3 days |
| Normal Direct Cost | ₹6,10,000 | — |
| Additional Crashing Cost | — | ₹22,000 |
| Crashed Direct Cost | — | ₹6,32,000 |
| Initial Critical Path | A-B-C-F-G | A-B-C-F-G |
| Target | — | 25 days |
| Status | Normal | Minimum direct cost for 25-day target under stated model |
27. What Does the Case Study Really Demonstrate?
The example demonstrates four different ideas.
27.1 CPM determines the normal schedule
The normal activity durations produce:
\[ \boxed{28\text{ days}} \]
with critical path:
\[ \boxed{A-B-C-F-G} \]
27.2 Float identifies noncritical flexibility
D and E each have:
\[ \boxed{3\text{ days of total float}} \]
27.3 Crashing reduces duration at additional direct cost
Three days are removed from the project duration at:
\[ \boxed{Rs. 22,000} \]
additional direct cost.
27.4 Optimization requires an objective
The phrase “optimal” has meaning only after specifying what is being optimized.
For example:
- minimum project duration;
- minimum direct cost;
- minimum total cost for a specified deadline;
- minimum cost subject to a deadline;
- time–cost–quality optimization.
In this case, the demonstrated optimization objective is:
Minimize additional direct crashing cost subject to completing the illustrative project within 25 days.
That objective makes the ₹22,000 result mathematically meaningful.
28. Why Crashing a Noncritical Activity May Not Shorten the Project
Suppose we reduce D:
\[ 5\rightarrow4 \]
D is on the second path:
\[ A-D-E-F-G \]
But the initial path duration is only 25 days.
The project is controlled by the 28-day path:
\[ A-B-C-F-G \]
Therefore, reducing D alone does not reduce the project completion time.
This is a fundamental CPM principle:
Reducing the duration of an activity is useful for project acceleration only if the activity lies on a path that controls the project completion time, or if the reduction changes the set of controlling paths.
This is also why construction time–cost optimization is more complicated than simply selecting the cheapest activity. Research on construction time–cost trade-offs explicitly considers resource-selection alternatives and the resulting combinatorial nature of the optimization problem.
29. Why the Critical Path Must Be Recalculated
The critical path is not permanently fixed.
If an activity on the critical path is shortened sufficiently, another path may become critical.
For example, suppose:
\[ Path\ 1=25\text{ days} \]
and:
\[ Path\ 2=25\text{ days} \]
Then both paths are critical.
If the manager crashes an activity belonging only to Path 1, Path 2 may continue to determine project completion.
Therefore, the correct procedure is:
\[ \boxed{ Crash \rightarrow Recalculate\ Network \rightarrow Identify\ Critical\ Path(s) \rightarrow Crash\ Again } \]
not:
\[ \text{Choose the cheapest activity repeatedly} \]
30. Direct Cost Is Not the Same as Total Project Cost
The ₹6,32,000 calculated above is the direct cost represented by the illustrative activity-cost data.
It should not automatically be called the project's total economic cost.
A real construction project may also have:
- site overhead;
- supervision costs;
- temporary facilities;
- financing costs;
- equipment-related costs;
- contractual penalties;
- delay costs;
- opportunity costs;
- quality-related costs.
Therefore:
\[ Total\ Project\ Cost \neq Direct\ Activity\ Cost \]
in general.
A simplified economic model might be:
\[ Total\ Cost = Direct\ Cost+Indirect\ Cost \]
If shortening the project reduces indirect costs, an accelerated schedule could sometimes be economically preferable even though direct crashing costs increase.
31. Linear Cost Slope is an Assumption
The cost-slope formula:
\[ \frac{Crash\ Cost-Normal\ Cost} {Normal\ Time-Crash\ Time} \]
is based on a simplified linear relationship between activity time and direct cost.
This assumption is useful for introductory CPM and time–cost calculations.
However, real construction resource choices may not follow a single linear relationship.
For example, reducing an activity from 10 to 9 days might require relatively little additional resource, whereas reducing it from 6 to 5 days might require a much larger intervention.
Research has therefore considered discrete and more sophisticated construction time–cost trade-off models rather than relying exclusively on simple linear cost slopes. Liu, Burns and Feng, for example, formulated construction time–cost trade-off as an LP/IP resource-selection problem.
32. Resource Constraints
The illustrative model does not explicitly include limited labour, equipment or material resources.
This is an important simplification.
In a real project, two activities may require the same crane, specialist crew or equipment at the same time.
Therefore, a schedule that is feasible from a pure precedence perspective may not be feasible from a resource perspective.
Construction research has explicitly incorporated resource-selection decisions into time–cost trade-off models. Liu, Burns and Feng describe choices involving crew size, equipment, construction methods and technologies and formulate an optimization procedure to select among alternatives.
Therefore, the illustrative model should be understood as a precedence-based CPM and direct-cost crashing model, not as a complete resource-constrained construction scheduling model.
33. Deterministic Activity Durations
The illustrative CPM model uses specified activity durations.
In actual construction, durations may be affected by:
- weather;
- material availability;
- labour productivity;
- equipment breakdown;
- site conditions;
- design changes;
- regulatory issues;
- supply-chain disruptions.
Therefore, actual activity durations may differ from planning estimates.
This is one reason why CPM and PERT are conceptually different. CPM is traditionally associated with specified/deterministic activity durations, whereas PERT was developed for situations involving uncertainty in activity durations.
34. CPM, PERT and Time–Cost Trade-Off
| Feature | CPM | PERT | Time–Cost Trade-Off |
|---|---|---|---|
| Main purpose | Project scheduling | Scheduling under uncertain activity times | Select time/resource/cost alternatives |
| Activity duration | Specified/deterministic in classical form | Probabilistic estimates | Alternative durations |
| Critical path | Yes | Yes | Usually part of the analysis |
| Float analysis | Yes | Yes | Yes, depending on model |
| Crashing | Commonly associated | Not its primary purpose | Central concept |
| Cost consideration | Can be incorporated | Not its primary focus | Central |
| Resource alternatives | Not inherently complete | Not inherently complete | Can be explicitly modelled |
Standard Operations Research literature treats project scheduling and time–cost trade-off as related but distinct modelling issues. Hillier and Lieberman's text, for example, includes a dedicated network model for optimizing a project's time–cost trade-off.
35. Evidence from Peer-Reviewed Construction Research
The theoretical distinction between CPM scheduling and broader construction optimization is supported by peer-reviewed research.
35.1 Liu, Burns and Feng (1995)
Liu, Burns and Feng published “Construction Time-Cost Trade-Off Analysis Using LP/IP Hybrid Method” in the Journal of Construction Engineering and Management, Volume 121, Issue 4, pages 446–454.
The paper describes construction planning as a resource-selection problem involving crew size, equipment, methods and technologies. It uses linear and integer programming to efficiently search for resource selections that optimize construction time and cost.
This is important because it demonstrates that advanced construction time–cost optimization can go beyond the simple rule of selecting the lowest cost slope.
35.2 Burns, Liu and Feng (1996)
A related paper, “The LP/IP Hybrid Method for Construction Time-Cost Trade-Off Analysis,” was published in Construction Management and Economics, Volume 14, Issue 3, pages 265–276.
The authors again address the selection of resources, crew sizes, equipment, methods and technologies and formulate an LP/IP approach for optimizing construction time and cost.
35.3 Khang and Myint (1999)
Khang and Myint published “Time, Cost and Quality Trade-Off in Project Management: A Case Study” in the International Journal of Project Management, Volume 17, Issue 4, pages 249–256.
Importantly, the study applied a time–cost–quality framework to an actual cement-factory construction project in Thailand. The researchers also discussed practical problems associated with estimating model parameters from real industrial data.
This provides a useful real-world contrast to the illustrative example used in this article.
What the Published Research Tells Us
The research literature suggests an important hierarchy:
Basic level
\[ \text{Network} \rightarrow \text{CPM} \rightarrow \text{Critical Path} \]
Intermediate level
\[ \text{CPM} \rightarrow \text{Crashing} \rightarrow \text{Time–Cost Trade-Off} \]
Advanced level
\[ \text{Time} + \text{Cost} + \text{Resources} + \text{Quality} + \text{Other Constraints} \rightarrow \text{Mathematical Optimization} \]
This progression is reflected in both standard Operations Research textbooks and construction research.
Limitations of the Illustrative Case Study
The numerical case deliberately uses a simplified model.
Its limitations include:
1. Illustrative data
The activity times and costs are constructed for educational purposes.
2. No real-site validation
The results have not been validated against measurements from an actual construction project.
3. Linear cost slopes
The model assumes a linear relationship between time reduction and additional direct cost.
4. Integer-day reductions
The example treats reductions in whole days.
5. No resource conflicts
Labour, equipment and material constraints are not explicitly modelled.
6. No uncertainty
Activity durations are treated as specified values.
7. No quality objective
The model does not mathematically represent the effect of acceleration on quality.
8. No indirect-cost model
The ₹6,32,000 figure represents direct activity cost under the illustrative assumptions, not a complete total-cost model.
These limitations are important because actual construction time–cost optimization can involve resource selection, quality and other practical considerations. Published research using real construction projects has demonstrated that parameter estimation and practical implementation can present significant challenges.
Practical Procedure for Applying CPM and Crashing
For a real construction project, the following procedure is appropriate as a basic CPM/time–cost framework.
Step 1: Identify project activities
Prepare a complete activity list.
Step 2: Determine precedence relationships
Identify the immediate predecessors of each activity.
Step 3: Estimate normal durations
Determine the planned duration of every activity.
Step 4: Construct the project network
Represent the precedence relationships.
Step 5: Perform the forward pass
Calculate:
\[ ES,\ EF \]
Step 6: Perform the backward pass
Calculate:
\[ LS,\ LF \]
Step 7: Calculate float
Use:
\[ TF=LS-ES \]
Step 8: Identify the critical path
Identify the zero-total-float path or paths.
Step 9: Identify feasible acceleration alternatives
For relevant activities, determine:
- crash duration;
- crash cost;
- resource requirement;
- practical feasibility.
Step 10: Calculate cost slopes where appropriate
\[ Cost\ Slope= \frac{Crash\ Cost-Normal\ Cost} {Normal\ Time-Crash\ Time} \]
Step 11: Select an acceleration decision
Consider the current critical path and the stated optimization objective.
Step 12: Recalculate the network
After each relevant change, recalculate the schedule.
Step 13: Check for multiple critical paths
Do not assume that the original critical path remains the only controlling path.
Step 14: Continue until the objective is achieved
The stopping condition might be:
- a contractual deadline;
- a desired completion time;
- minimum direct cost for a target duration;
- minimum total cost;
- or another specified objective.
Key Lessons from the Case Study
Lesson 1: The longest path determines project duration
The normal project duration is 28 days, not the sum of all activity durations.
Lesson 2: Critical activities have zero total float
A, B, C, F and G are critical in the normal schedule.
Lesson 3: Float provides flexibility
D and E each have 3 days of total float.
Lesson 4: Crashing is not the same as simply choosing the cheapest activity
The activity must be relevant to the controlling path.
Lesson 5: The network must be recalculated
Every significant duration change can alter the critical path.
Lesson 6: The shortest schedule and cheapest schedule are different concepts
A project can be accelerated at increasing direct cost.
Lesson 7: An optimization statement requires a defined objective
In this example, the objective is:
Minimize additional direct crashing cost subject to a 25-day project-duration target.
Lesson 8: Real construction optimization is more complicated
Resource constraints, uncertainty, quality and nonlinear cost relationships may require more advanced models.
Final Results
The illustrative project produces the following results:
\[ \boxed{Normal\ Duration=28\ days} \]
\[ \boxed{Normal\ Direct\ Cost=Rs. 6,10,000} \]
\[ \boxed{Critical\ Path=A-B-C-F-G} \]
After the selected crashing decisions:
\[ A:4\rightarrow3 \]
\[ G:3\rightarrow2 \]
\[ B:6\rightarrow5 \]
the project duration becomes:
\[ \boxed{25\ days} \]
Additional direct crashing cost:
\[ \boxed{Rs. 22,000} \]
Crashed direct cost:
\[ \boxed{Rs. 6,32,000} \]
Time saved:
\[ 28-25=\boxed{3\ days} \]
Under the stated illustrative assumptions, ₹22,000 is the minimum additional direct cost for achieving the 25-day target.
Conclusion
The Critical Path Method is a fundamental tool for construction project scheduling because it provides a systematic way to represent activity relationships, calculate project duration, determine float and identify the activities that control project completion.
In the illustrative construction project presented here, the normal schedule requires 28 days, with A–B–C–F–G as the critical path. The second path, A–D–E–F–G, requires 25 days and therefore has three days of difference from the controlling path under the normal schedule.
The time–cost analysis demonstrates that the project can be reduced from 28 to 25 days by crashing A, G and B. The additional direct cost is ₹22,000, producing a direct cost of ₹6,32,000.
More importantly, the example demonstrates a methodological principle:
CPM identifies what controls project duration; crashing and time–cost analysis determine how duration may be changed at additional cost.
For the specified 25-day target, the illustrative model shows that the ₹22,000 additional direct cost is the minimum among the available integer-day crashing alternatives under the stated linear-cost assumptions.
However, this conclusion should not be generalized to all construction projects. Real construction projects may involve resource limitations, uncertain activity durations, nonlinear or discrete cost relationships, quality considerations and other constraints. Peer-reviewed construction research has therefore developed mathematical programming and other optimization approaches for more complex time–cost problems.
Thus, CPM should be viewed as a foundation for construction scheduling and time–cost analysis, rather than as a complete solution to every real-world construction optimization problem.
Frequently Asked Questions
Q. What is CPM in construction project management?
CPM is a deterministic network-based scheduling technique used to calculate project duration, identify critical activities and determine activity float.
Q. What is the critical path?
The critical path is the path through the project network that determines the project completion time for the specified activity durations.
Q. What is crashing in CPM?
Crashing is the process of reducing activity duration through feasible acceleration alternatives, generally at additional direct cost.
Q. What is cost slope?
Cost slope is the additional direct cost per unit of activity-duration reduction under the conventional linear time–cost model:
\[ Cost\ Slope= \frac{Crash\ Cost-Normal\ Cost} {Normal\ Time-Crash\ Time} \]
Q. Does CPM itself optimize project cost?
Not necessarily. CPM determines the schedule for specified durations. Cost optimization requires an explicit objective and additional cost/resource information.
Q. Why should the critical path be recalculated after crashing?
Because reducing a critical activity can make another path equally long or longer relative to the previously critical path.
Q. Is the case study in this article real?
No. The numerical construction project is explicitly illustrative. It was created to demonstrate the CPM and crashing methodology.
Q. Is ₹6,32,000 the total cost of the construction project?
No. It is the direct activity cost represented by the illustrative data after the specified crashing decisions. A real project's total cost may include indirect and other costs.
43. References
Foundational CPM Source
Kelley, J. E., Jr., & Walker, M. R. (1959). Critical-path planning and scheduling. IRE-AIEE-ACM '59 (Eastern): Papers presented at the December 1–3, 1959, Eastern Joint IRE-AIEE-ACM Computer Conference, 160–173. DOI: 10.1145/1460299.1460318.
Standard Operations Research Textbooks
Hillier, F. S., & Lieberman, G. J. Introduction to Operations Research. McGraw-Hill. The text includes project management with PERT/CPM and a network model for optimizing a project's time–cost trade-off.
Taha, H. A. (2022). Operations Research: An Introduction, 11th ed. Pearson. Pearson describes the book as combining theory, applications and computation and emphasizing practical implementation of Operations Research techniques.
Peer-Reviewed Construction and Project-Management Research
Liu, L., Burns, S. A., & Feng, C.-W. (1995). Construction time-cost trade-off analysis using LP/IP hybrid method. Journal of Construction Engineering and Management, 121(4), 446–454. DOI: 10.1061/(ASCE)0733-9364(1995)121:4(446).
Burns, S. A., Liu, L., & Feng, C.-W. (1996). The LP/IP hybrid method for construction time-cost trade-off analysis. Construction Management and Economics, 14(3), 265–276. DOI: 10.1080/014461996373511.
Khang, D. B., & Myint, Y. M. (1999). Time, cost and quality trade-off in project management: A case study. International Journal of Project Management, 17(4), 249–256. DOI: 10.1016/S0263-7863(98)00043-X. The study applies a time–cost–quality framework to an actual cement-factory construction project and discusses practical issues in applying such models to real industrial projects.
Note: Academic Status of This Article
This article should be classified as an:
Educational / Tutorial Case Study based on established CPM and time–cost trade-off methodology.
The numerical construction project is illustrative, not empirical.
Therefore, it should not be presented as an original real-world research study or as a substitute for a peer-reviewed empirical construction case study.
The theoretical framework is grounded in established Operations Research textbooks and published CPM/time–cost trade-off research, while the numerical example is independently constructed to make the methodology transparent and reproducible.
Comments
Post a Comment