Cocomo

The CoCoMo Model (The Constructive Cost Model)

Tom Rethard based on a presentation by Jim Paulson of Novatel Wireless

COCOMO Outline  

Introduction The Basic Model   

 

Equations When You Should Use It Limitations

Intermediate Model Detailed Model

CSE 4310/5392

2

COCOMO 





COCOMO is one of the most widely used software estimation models in the world It was developed by Barry Boehm in 1981 COCOMO predicts the effort and schedule for a software product development based on inputs relating to the size of the software and a number of cost drivers that affect

CSE 4310/5392

3

COCOMO: Three Models 

COCOMO has three different models that reflect the complexity:   

the Basic Model the Intermediate Model and the Detailed Model

CSE 4310/5392

4

The Development Modes: Project Characteristics 

Organic Mode   



Semidetached Mode 



developed in a familiar, stable environment, similar to the previously developed projects relatively small and requires little innovation intermediate between Organic and Embedded

Embedded Mode  

tight, inflexible constraints and interface requirements The product requires great innovation

CSE 4310/5392

5

COCOMO: Some Assumptions 





Primary cost driver is the number of Delivered Source Instructions (DSI) developed by the project COCOMO estimates assume that the project will enjoy good management by both the developer and the customer Assumes the requirements specification is not substantially changed after the plans and requirements phase CSE 4310/5392

6

Basic COCOMO Model 

Basic COCOMO model estimates the software development effort using only a single predictor variable (size in DSI) and three software development modes

CSE 4310/5392

7

Basic COCOMO Model: Equations Mode

Effort

Schedule

E=2.4*(KDSI)

1.05

Semidetached E=3.0*(KDSI)

1.12

Embedded

1.20

Organic

CSE 4310/5392

E=3.6*(KDSI)

TDEV=2.5*(E)

0.38

TDEV=2.5*(E)

0.35

TDEV=2.5*(E)

0.32

8

Basic COCOMO Model: When Should You Use It 

Basic COCOMO is good for quick, early, rough order of magnitude estimates of software costs

CSE 4310/5392

9

Basic COCOMO Model: Limitations 



Its accuracy is necessarily limited because of its lack of factors which have a significant influence on software costs The Basic COCOMO estimates are within a factor of 1.3 only 29% of the time, and within a factor of 2 only 60% of the time

CSE 4310/5392

10

Basic COCOMO Model: An Example 



  



We have determined our project fits the characteristics of Semi-Detached mode We estimate our project will have 32,000 Delivered Source Instructions. Using the formulas, we can estimate: Effort = 3.0*(32) 1.12 = 146 man-months Schedule = 2.5*(146) 0.35 = 14 months Productivity = 32,000 DSI / 146 MM = 219 DSI/MM Average Staffing = 146 MM /14 months CSE 4310/5392

= 10 FSP

11

Intermediate COCOMO Model 

The Intermediate Model estimates the software development effort by using fifteen cost driver variables besides the size variable used in Basic COCOMO

CSE 4310/5392

12

Intermediate COCOMO Model 

Four areas for drivers    

Product itself Computer Personnel Project itself

CSE 4310/5392

13

Intermediate Model: Cost Driver Categories 

Product Attributes 

 

RELY --- Required Software Reliability DATA --- Data Base Size CPLX --- Software Product Complexity

CSE 4310/5392

14

Intermediate Model: Cost Driver Categories 

Computer Attributes    

TIME --- Execution Time Constraint STOR --- Main Storage Constraint VIRT --- Virtual Machine Volatility TURN --- Computer Turnaround Time

CSE 4310/5392

15

Intermediate Model: Cost Driver Categories 

Personnel Attributes     

ACAP --- Analyst Capability AEXP --- Applications Experience PCAP --- Programmer Capability VEXP --- Virtual Machine Experience LEXP --- Programming Language Experience

CSE 4310/5392

16

Intermediate Model: Cost Driver Categories 

Project Attributes 

 

MODP --- Modern Programming Practices TOOL --- Use of Software Tools SCED --- Required Development Schedule

CSE 4310/5392

17

Intermediate Model: Effort Multipliers 

Table of Effort Multipliers for each of the Cost Drivers is provided with ranges depending on the ratings Very Very Extra

Cost Driver

CPLX Product Complexity CSE 4310/5392

Low Low Nom High High High

0.70 0.85 1.00 1.15 1.30 1.65 18

Intermediate Model: Equations Mode

Effort

Schedule

Organic

E=EAF*3.2*(KDSI)

1.05

1.12

TDEV=2.5*(E)

1.20

TDEV=2.5*(E)

SemiE=EAF*3.0*(KDSI) detached Embedded E=EAF*2.8*(KDSI) CSE 4310/5392

0.38

TDEV=2.5*(E)

0.35

0.32

19

Intermediate Model: When Should You Use It The Intermediate Model can be applied across the entire software product for easily and rough cost estimation during the early stage  or it can be applied at the software product component level for more accurate cost estimation in more detailed CSE 4310/5392 

20

Intermediate Model: Limitations 





The Intermediate Model estimates are within 20% of the actuals 68% of the time Its effort multipliers are phaseinsensitive It can be very tedious to use on a product with many components

CSE 4310/5392

21

Intermediate Model: An Example Project A is to be a 32,000 DSI semidetached software. It is in a mission critical area, so the reliability is high (RELY=high=1.15). Then we can estimate:  Effort = 1.15*3.0*(32)1.12 = 167 man-months  Schedule = 2.5*(167)0.35 = 15 months  Productivity = 32,000 DSI/167 MM CSE 4310/5392 = 192 DSI/MM 

22

Detailed COCOMO Model: How is it Different? 

Phase-sensitive Effort Multipliers The effort multipliers for every cost drivers are different during the software development phases



Module-Subsystem-System Hierarchy

The software product is estimated in the three level hierarchical decomposition. The fifteen cost drivers are related to module or subsystem level CSE 4310/5392

23

Detailed COCOMO Model: Module-Subsystem-System 

Module level  



Subsystem Level 





cost drivers tend to vary at the lowest level CPLX, PCAP, VEXP, LEXP cost drivers tend to vary from subsystem to subsystem, but are the same for modules in a subsystem RELY, DATA, TIME, STOR, VIRT

System Level 

overall project relations such as nominal effort and schedule equations

CSE 4310/5392

24

Detailed COCOMO Model: Equations 



Detailed Model uses the same equations for estimations as the Intermediate Model Detailed Model uses a very complex procedure to calculate estimation. The procedure uses the DSIs for subsystems and modules, and module level and subsystem level effort multipliers as inputs

CSE 4310/5392

25

Detailed COCOMO Model: When Should You Use It? 



The Detailed Model can estimate the staffing, cost, and duration of each of the development phases, subsystems, modules It allows you to experiment with different development strategies, to find the plan that best suits your needs and resources

CSE 4310/5392

26

Detailed COCOMO Model: Limitations 



Requires substantially more time and effort to calculate estimates than previous models Detailed Model estimates are within 20% of the actuals 70% of the time

CSE 4310/5392

27

Summary 





Cost estimation models are not a substitute for a detailed estimate by task by project management Cost estimation models highly depend on the user’s knowledge of application domain, analysis ability and the understanding of the cost model itself The process of meeting the estimate is equally critical to the process of making the estimate

CSE 4310/5392

28

Links 

COCOMO II web page can be found at: 

sunset.usc.edu/research/COCOMOII/index.html

CSE 4310/5392

29

Questions / Discussion 



Issues to discuss: Do you think cost multipliers are project phase dependent like COCOMO’s detailed model suggests? Is cost estimation only suitable for the Waterfall lifecycle, or can it be used with other life cycles, such as incremental? CSE 4310/5392

30

Questions / Discussion 



Issues to discuss: Your cost estimate uses three methods which return widely varying results? How do you determine which method is right? What methods or models have you seen used in the workplace? How successful were they? CSE 4310/5392

31

Questions / Discussion 

Issues to discuss: Is it possible to estimate the size using soft requirements?

CSE 4310/5392

32