ch26 | Top Down And Bottom Up Design | Profit (Accounting)

Software cost estimation

©Ian Sommerville 2004

Software Engineering, 7th edition. Chapter 26

Slide 1

Objectives    

To introduce the fundamentals of software costing and pricing To describe three metrics for software productivity assessment To explain why different techniques should be used for software estimation To describe the principles of the COCOMO 2 algorithmic cost estimation model

©Ian Sommerville 2004

Software Engineering, 7th edition. Chapter 26

Slide 2

Topics covered 
  

Software productivity Estimation techniques Algorithmic cost modelling Project duration and staffing

©Ian Sommerville 2004

Software Engineering, 7th edition. Chapter 26

Slide 3

Fundamental estimation questions   



How much effort is required to complete an activity? How much calendar time is needed to complete an activity? What is the total cost of an activity? Project estimation and scheduling are interleaved management activities.

©Ian Sommerville 2004

Software Engineering, 7th edition. Chapter 26

Slide 4

Software cost components 
 

Hardware and software costs. Travel and training costs. Effort costs (the dominant factor in most projects)
‡ ‡ The salaries of engineers involved in the project; Social and insurance costs. Costs of building, heating, lighting. Costs of networking and communications. Costs of shared facilities (e.g library, staff restaurant, etc.). 

Effort costs must take overheads into account
‡ ‡ ‡

©Ian Sommerville 2004

Software Engineering, 7th edition. Chapter 26

Slide 5

Chapter 26 Slide 6 ©Ian Sommerville 2004 .Costing and pricing    Estimates are made to discover the cost. to the developer. political and business considerations influence the price charged. There is not a simple relationship between the development cost and the price charged to the customer. economic. Software Engineering. 7th edition. of producing a software system. Broader organisational.

Cost estimate uncertainty Contractual terms Requirements volatility Financial health ©Ian Sommerville 2004 Software Engineering. 7th edition. Chapter 26 Slide 7 . high prices can be charged for changes to the requirements. an organisation may lower its price to win a contract. The price charged may then be less than if the software source code is handed over to the customer. it may increase its price by some contingency over and above its normal profit. Accepting a low profit on one project may give the opportunity of more profit later. If an organisation is unsure of its cost estimate. The experience gained may allow new products to be developed. If the requirements are likely to change. A c ustomer may be willing to allow the developer to retain ownership of the source code and reuse it in other projects. Developers in financial difficulty may lower their price to ga in a c ontract.Software pricing factors Market opportunity A d evelopment organisation may quote a low price because it wishes to move into a new segment of the software market. It is better to make a sm aller than normal profit or break even than to go out of business. After the contract is awarded.

Essentially. Not quality-oriented although quality assurance is a factor in productivity assessment. 7th edition. Software Engineering.Software productivity    A measure of the rate at which individual engineers involved in software development produce software and associated documentation. Chapter 26 Slide 8 ©Ian Sommerville 2004 . we want to measure useful functionality produced per time unit.

Function-related measures based on an estimate of the functionality of the delivered software. Function-points are the best known of this type of measure.Productivity measures   Size related measures based on some output from the software process. object code instructions. 7th edition. Chapter 26 Slide 9 . ©Ian Sommerville 2004 Software Engineering. etc. This may be lines of delivered source code.

how many function points). Chapter 26 Slide 10 . ©Ian Sommerville 2004 Software Engineering. Estimating contractor productivity (e. Estimating the total number of programmer months that have elapsed.g.Measurement problems    Estimating the size of the measure (e. 7th edition.g. documentation team) and incorporating this estimate in overall estimate.

  What programs should be counted as part of the system? This model assumes that there is a linear relationship between system size and volume of documentation. ©Ian Sommerville 2004 Software Engineering. How does this correspond to statements as in Java which can span several lines or where there can be several statements on one line. 7th edition. Chapter 26 Slide 11 .Lines of code  What's a line of code? ‡ ‡ The measure was first proposed when programs were typed on cards with one line per card.

Productivity comparisons  The lower level the language. ©Ian Sommerville 2004 Software Engineering. the more productive the programmer ‡ The same functionality takes more code to implement in a lower-level language than in a high-level language.  The more verbose the programmer. Chapter 26 Slide 12 . the higher the productivity ‡ Measures of productivity based on lines of code suggest that programmers who write verbose code are more productive than programmers who write compact code. 7th edition.

7th edition.System development times Analysis Assembly code High-level language 3 weeks 3 weeks Design 5 weeks 5 weeks Co ing 8 weeks 4 weeks Testing 10 weeks 6 weeks Pro uctivity Documentation 2 weeks 2 weeks Size Assembly code High-level language 5000 lines 1500 lines ffort 28 weeks 20 weeks 714 lines/month 300 lines/month ©Ian Sommerville 2004 Software Engineering. Chapter 26 Slide 13 .

UFC =§ (numb of elements ofgiven type) v(weight) er ©Ian Sommerville 2004 Software Engineering. external interfaces.Function points  Based on a combination of program characteristics ‡ ‡ ‡ ‡ external inputs and outputs. user interactions. Chapter 26 Slide 14 . files used by the system.  A weight is associated with each of these and the function point count is computed by multiplying each raw count by the weight and summing all values. 7th edition.

Function points   The function point count is modified by complexity of the project FPs can be used to estimate LOC depending on the average number of LOC per FP for a given language ‡ ‡ LOC = AVC * number of function points. AVC is a language-dependent factor varying from 200300 for assemble language to 2-40 for a 4GL. Chapter 26 Slide 15 . 7th edition. ©Ian Sommerville 2004 Software Engineering.  FPs are very subjective. They depend on the estimator ‡ Automatic function-point counting is impossible.

7th edition. The number of object points in a program is a weighted estimate of ‡ ‡ ‡ The number of separate screens that are displayed. Object points are NOT the same as object classes. The number of program modules that must be developed to supplement the database code. Chapter 26 Slide 16 . The number of reports that are produced by the system.Object points    Object points (alternatively named application points) are an alternative function-related measure to function points when 4Gls or similar languages are used for development. ©Ian Sommerville 2004 Software Engineering.

Object point estimation    Object points are easier to estimate from a specification than function points as they are simply concerned with screens. They can therefore be estimated at a fairly early point in the development process. 7th edition. ©Ian Sommerville 2004 Software Engineering. At this stage. it is very difficult to estimate the number of lines of code in a system. reports and programming language modules. Chapter 26 Slide 17 .

Systems programs . 150-400 LOC/P-month. productivity has been measured between 4 and 50 object points/month depending on tool support and developer capability. 40-160 LOC/P-month. ©Ian Sommerville 2004 Software Engineering. In object points. 200-900 LOC/P-month.Productivity estimates     Real-time embedded systems. 7th edition. Chapter 26 Slide 18 . Commercial applications.

Chapter 26 Slide 19 . . f C . . 7th edition. f ff y f ff ©Ian Sommerville 2004 Software Engineering. L y y W G y A w C . y. f w y y. w y.Factors affecting productivity A K w fw f . f w . CA E . y . j . E y P P j z .

If requirements are constantly changing then an approach based on counting lines of code is not meaningful as the program itself is not static. 7th edition. Chapter 26 Slide 20 . It is not clear how productivity/quality metrics are related. ©Ian Sommerville 2004 Software Engineering. Productivity may generally be increased at the cost of quality.Quality and productivity     All metrics based on volume/unit time are flawed because they do not take quality into account.

The software may run on unfamiliar computers or use new technology. The people in the project may be unknown.  Project cost estimates may be self-fulfilling ‡ ©Ian Sommerville 2004 Software Engineering. Chapter 26 Slide 21 .Estimation techniques  There is no simple way to make an accurate estimate of the effort required to develop a software system ‡ ‡ ‡ Initial estimates are based on inadequate information in a user requirements definition. The estimate defines the budget and the product is adjusted to meet the budget. 7th edition.

Development for and with reuse. The use of CASE tools and program generators. 7th edition. Chapter 26 Slide 22 . ©Ian Sommerville 2004 Software Engineering.Changing technologies  Changing technologies may mean that previous estimating experience does not carry over to new systems ‡ ‡ ‡ ‡ ‡ ‡ ‡ Distributed object systems rather than mainframe systems. Development using scripting languages. Use of off-the-shelf software. Use of ERP or database-centred systems. Use of web services.

Estimation techniques      Algorithmic cost modelling. Parkinson's Law. Expert judgement. 7th edition. ©Ian Sommerville 2004 Software Engineering. Estimation by analogy. Pricing to win. Chapter 26 Slide 23 .

The cost o a ne pro ect is estimated by analogy ith these completed pro ects. the e ort required is estimated to be 60 personmonths. I the so t are has to be delivered in 12 months and 5 people are available. They each estimate the pro ect cost. The estimated e ort depends on the customer s b udget and not on the so t are unctionality.Estimation techniques Algorithmic cost modelling xpert udgement A m odel based on historical cost in ormation that relates some so t are metric (usually its si e) to the pro ect cost is used. Myers (Myers 1989) gives a very clear description o this approach. stimation by analogy arkinson s a ricing to in ©Ian Sommerville 2004 Software Engineering. The cost is determined by available resources rather than by ob ective assessment. Chapter 26 Slide 24 . This technique is applicable hen other pro ects in the same application domain have been completed. The so t are cost is estimated to be hatever the customer has available to spend on the pro ect. arkinson s a states that ork expands to ill the time available. The estimation process iterates until an agreed estimate is reached. These estimates are compared and discussed. Several experts on the proposed so t are development techniques and the application domain are consulted. An estimate is m ade o that metric and the model predicts the e ort required. 7th edition.

The probability that the customer gets the system he or she wants is small. 7th edition.  Disadvantages: ‡ ©Ian Sommerville 2004 Software Engineering. Costs do not accurately reflect the work required. Chapter 26 Slide 25 .Pricing to win   The project costs whatever the customer has to spend on it. Advantages: ‡ You get the contract.

7th edition. Add these efforts to reach a final estimate.  Bottom-up ‡ ©Ian Sommerville 2004 Software Engineering.Top-down and bottom-up estimation   Any of these approaches may be used topdown or bottom-up. Chapter 26 Slide 26 . Top-down ‡ Start at the system level and assess the overall system functionality and how this is delivered through sub-systems. Start at the component level and estimate the effort required for each component.

7th edition. Chapter 26 Slide 27 . configuration management and documentation. Can underestimate the cost of solving difficult low-level technical problems. Takes into account costs such as integration. ©Ian Sommerville 2004 Software Engineering.Top-down estimation    Usable without knowledge of the system architecture and the components that might be part of the system.

Bottom-up estimation    Usable when the architecture of the system is known and components identified. It may underestimate the costs of system level activities such as integration and documentation. Chapter 26 Slide 28 . 7th edition. This can be an accurate method if the system has been designed in detail. ©Ian Sommerville 2004 Software Engineering.

7th edition. Pricing to win is sometimes the only applicable method.Estimation methods      Each method has strengths and weaknesses. Chapter 26 Slide 29 . then you have insufficient information available to make an estimate. If these do not return approximately the same result. Some action should be taken to find out more in order to make more accurate estimates. ©Ian Sommerville 2004 Software Engineering. Estimation should be based on several methods.

Chapter 26 Slide 30 . when detailed information is lacking it may be the only appropriate strategy. ©Ian Sommerville 2004 Software Engineering. The project cost is agreed on the basis of an outline proposal and the development is constrained by that cost. However.Pricing to win     This approach may seem unethical and unbusinesslike. A detailed specification may be negotiated or an evolutionary approach used for system development. 7th edition.

  The most commonly used product attribute for cost estimation is code size. B reflects the disproportionate effort for large projects and M is a multiplier reflecting product. Chapter 26 . process and people attributes. 7th edition. Slide 31 ©Ian Sommerville 2004 Software Engineering. Most models are similar but they use different values for A. project and process attributes whose values are estimated by project managers: ‡ ‡ Effort = A v SizeB v M A is an organisation-dependent constant. B and M.Algorithmic cost modelling  Cost is estimated as a mathematical function of product.

Distribution of system. ©Ian Sommerville 2004 Software Engineering. Several factors influence the final size ‡ ‡ ‡ Use of COTS and components.Estimation accuracy   The size of a software system can only be known accurately when it is finished. 7th edition.  As the development process progresses then the size estimate becomes more accurate. Chapter 26 Slide 32 . Programming language.

Chapter 26 Slide 33 . 7th edition.Estimate uncertainty ©Ian Sommerville 2004 Software Engineering.

Well-documented. µindependent¶ model which is not tied to a specific software vendor. Chapter 26 Slide 34 . etc. reuse.The COCOMO model     An empirical model based on project experience. ©Ian Sommerville 2004 Software Engineering. Long history from initial version published in 1981 (COCOMO-81) through various instantiations to COCOMO 2. 7th edition. COCOMO 2 takes into account different approaches to software development.

regulations and operational procedures. software.1 2 vM Descri tion Well-understood applications developed by small teams. Chapter 26 Slide 35 . Complex projects where the software is part of a strongly coupled complex of hardware. More complex projects where team members may have limited experience of related systems.0 5 v M PM = 3.6 (KDS I) 1 .2 0 v M ©Ian Sommerville 2004 Software Engineering. 7th edition.0 (KDS I) 1 . Embedded PM = 3.4 (KDS I) 1 .COCOMO 81 Project com lexity Simple Moderate Formula PM = 2.

there have been many changes in software engineering practice and COCOMO 2 is designed to accommodate different approaches to software development. ©Ian Sommerville 2004 Software Engineering. 7th edition. Chapter 26 Slide 36 .COCOMO 2   COCOMO 81 was developed with the assumption that a waterfall process would be used and that all software would be developed from scratch. Since its formulation.

Post-architecture model. Reuse model. Chapter 26 Slide 37 . Used when software is composed from existing parts. Used to compute the effort of integrating reusable components. Used once the system architecture has been designed and more information about the system is available. ©Ian Sommerville 2004 Software Engineering.COCOMO 2 models   COCOMO 2 incorporates a range of sub-models that produce increasingly detailed software estimates. 7th edition. Early design model. The sub-models in COCOMO 2 are: ‡ ‡ ‡ ‡ Application composition model. Used when requirements are available but design has not yet started.

Use of COCOMO 2 models ©Ian Sommerville 2004 Software Engineering. Chapter 26 Slide 38 . 7th edition.

Application composition model     Supports prototyping projects and projects where there is extensive reuse.%reuse/100 ) ) / PROD PM is the effort in person-months. NAP is the number of application points and PROD is the productivity. 7th edition. Chapter 26 Slide 39 . Based on standard estimates of developer productivity in application (object) points/month. Formula is ‡ ‡ PM = ( NAP v (1 . Takes CASE tool use into account. ©Ian Sommerville 2004 Software Engineering.

Object point productivity Developer s experience and capability ICASE maturity and capability PROD (NOP/month) Very low Very low 4 Low Low 7 Nominal Nominal 13 High High 25 Very high Very high 50 ©Ian Sommerville 2004 Software Engineering. 7th edition. Chapter 26 Slide 40 .

Size in KLOC. 7th edition.1 to 1.94 in initial calibration. Software Engineering.Early design model   Estimates can be made after the requirements have been agreed. Chapter 26 Slide 41 ©Ian Sommerville 2004 . B varies from 1.24 depending on novelty of the project. risk management approaches and the process maturity. development flexibility. Based on a standard formula for algorithmic models ‡ ‡ ‡ PM = A v SizeB v M where M = PERS v RCPX v RUSE v PDIF v PREX v FCIL v SCED. A = 2.

PDIF . the non-functional requirements. PREX . PERS . the familiarity with the development platform.the team support facilities. RUSE . 7th edition. etc.product reliability and complexity.required schedule. Chapter 26 Slide 42 ©Ian Sommerville 2004 .the reuse required.personnel experience. FCIL .platform difficulty. Software Engineering.personnel capability.Multipliers  Multipliers reflect the capability of the developers. SCED . ‡ ‡ ‡ ‡ ‡ ‡ ‡ RCPX .

White-box reuse where code is modified. There are two versions: ‡ ‡ Black-box reuse where code is not modified. Chapter 26 Slide 43 . An effort estimate (PM) is computed. A size estimate equivalent to the number of lines of new source code is computed. 7th edition. This then adjusts the size estimate for new code.The reuse model   Takes into account black-box code that is reused without change and code that has to be adapted to integrate it with new code. ©Ian Sommerville 2004 Software Engineering.

©Ian Sommerville 2004 Software Engineering. Chapter 26 Slide 44 .Reuse model estimates 1  For generated code: ‡ ‡ ‡ ‡ PM = (ASLOC * AT/100)/ATPROD ASLOC is the number of lines of generated code AT is the percentage of code automatically generated. ATPROD is the productivity of engineers in integrating this code. 7th edition.

Chapter 26 Slide 45 . ©Ian Sommerville 2004 Software Engineering. the costs of understanding how to integrate the code and the costs of reuse decision making. 7th edition.Reuse model estimates 2  When code has to be understood and integrated: ‡ ‡ ‡ ESLOC = ASLOC * (1-AT/100) * AAM. ASLOC and AT as before. AAM is the adaptation adjustment multiplier computed from the costs of changing the reused code.

Chapter 26 Slide 46 . An estimate of the number of lines of code that have to be modified according to requirements changes. ©Ian Sommerville 2004 Software Engineering. The code size is estimated as: ‡ ‡ ‡ Number of lines of new code to be developed. Estimate of equivalent number of lines of new code computed using the reuse model. 7th edition.Post-architecture level   Uses the same formula as the early design model but with 17 rather than 7 associated multipliers.

7th edition. ‡ ‡ ‡ ‡ ‡ Precedenteness .Very high (1) Architecture/risk resolution .new team .The exponent term   This depends on 5 scale factors (see next slide). Chapter 26 Slide 47 ©Ian Sommerville 2004 . The company has a CMM level 2 rating.new project (4) Development flexibility .some control . Low .nominal (3) Process maturity .01 A company takes on a project in a new domain. The client has not defined the process to be used and has not allowed time for risk analysis.17. Software Engineering.nominal (3)  Scale factor is therefore 1.no client involvement .(5) Team cohesion . Their sum/100 is added to 1.V.No risk analysis .

Very low means little analysis. The computation of this value depends on the CMM Maturity Questionnaire but an estimate can be achieved by subtracting the CMM process maturity level from 5. 7th edition. Chapter 26 Slide 48 . Extra high means that the organisation is completely familiar with this application domain. Reflects the extent of risk analysis carried out. Very low means very difficult interactions. Extra high means a complete a thorough risk analysis. Extra high means an integrated and effective team with no communication problems. Reflects the degree of flexibility in the development process. Very low means a prescribed process is used. Very low means no previous experience.Exponent scale factors Precedentedness Reflects the previous experience of the organisation with this type of project. Reflects how well the development team know each other and work together. Extra high means that the client only sets general goals. Reflects the process maturity of the organisation. Development flexibility Architecture/risk resolution Team cohesion Process maturity ©Ian Sommerville 2004 Software Engineering.

Software Engineering. 7th edition.  Computer attributes ‡  Personnel attributes ‡ Multipliers that take the experience and capabilities of the people working on the project into account. Chapter 26 Slide 49  Project attributes ‡ ©Ian Sommerville 2004 .Multipliers  Product attributes ‡ Concerned with required characteristics of the software product being developed. Concerned with the particular characteristics of the software development project. Constraints imposed on the software by the hardware platform.

75 None.17 System si e (including factors for reuse 128.21 Low. Chapter 26 Slide 50 .72 Normal. multiplier = 0.29 2306 erson-mont s Very low. multiplier = 1. multiplier = 1 Very high. multiplier = 0. multiplier = 1 295 erson-mont s ©Ian Sommerville 2004 Software Engineering.Effects of cost drivers Exponent value 1. multiplier = 1. 7th edition.75 Very low. 000 DSI and requirements volatility) Initial COCOMO estimate it out 730 erson-mont s cost rivers Reliability Complexity Memory constraint Tool use Schedule Adjusted COCOMO estimate Reliability Complexity Memory constraint Tool use Schedule Adjusted COCOMO estimate Very high. multiplier = 1.12 Accelerated. multiplier = 1. multiplier = 0.39 Very high.3 High. multiplier = 1.

Embedded spacecraft system ‡ ‡ ‡ Must be reliable. Target hardware. Development platform. Development effort.  Cost components ‡ ‡ ‡ ©Ian Sommerville 2004 Software Engineering. 7th edition.Project planning   Algorithmic cost models provide a basis for project planning as they allow alternative strategies to be compared. Must minimise weight (number of chips). Multipliers on reliability and computer constraints > 1. Chapter 26 Slide 51 .

Management options ©Ian Sommerville 2004 Software Engineering. 7th edition. Chapter 26 Slide 52 .

06 1 1 1.06 1 1 1.39 1. Chapter 26 Slide 53 .39 1.22 0.39 1.86 0. 7th edition.39 1.72 1.11 1.11 1 1 0.86 1.39 1.Management option costs Option RELY STOR TIME TOOLS LTEX Total effort Soft are cost A B C D E F 1.12 0.39 1.12 1 1.22 1 0.84 63 88 60 51 56 57 949393 1313550 895653 769008 844425 851180 Hardware cost 100000 Total cost 120000 105000 100000 220000 120000 1049393 1402025 1000653 897490 1044159 1002706 ©Ian Sommerville 2004 Software Engineering.86 0.11 1 1.84 1.

Option choice  Option D (use more experienced staff) appears to be the best alternative ‡ However. the model reveals the importance of staff experience in software development.   Option C (upgrade memory) has a lower cost saving but very low risk. Overall. ©Ian Sommerville 2004 Software Engineering. it has a high associated risk as experienced staff may be difficult to find. 7th edition. Chapter 26 Slide 54 .

Calendar time can be estimated using a COCOMO 2 formula ‡ ‡ TDEV = 3 v (PM)(0.Project duration and staffing   As well as effort estimation.2*(B-1. managers must estimate the calendar time required to complete a project and when staff will be required.  The time required is independent of the number of people working on the project. ©Ian Sommerville 2004 Software Engineering.01)) PM is the effort computation and B is the exponent computed as discussed above (B is 1 for the early prototyping model). 7th edition. This computation predicts the nominal schedule for the project.33+0. Chapter 26 Slide 55 .

The number of people working on a project varies depending on the phase of the project. Chapter 26 Slide 56 .Staffing requirements     Staff required can¶t be computed by diving the development time by the required schedule. the more total effort is usually required. A very rapid build-up of people often correlates with schedule slippage. 7th edition. The more people who work on the project. ©Ian Sommerville 2004 Software Engineering.

domain experience. Software may be priced to gain a contract and the functionality adjusted to the price. the development project.Key points    There is not a simple relationship between the price charged for a system and its development costs. Factors affecting productivity include individual aptitude. Software Engineering. tool support and the working environment. 7th edition. the project size. Chapter 26 Slide 57 ©Ian Sommerville 2004 .

Key points     Different techniques of cost estimation should be used when estimating costs. Algorithmic cost models support quantitative option analysis as they allow the costs of different options to be compared. Chapter 26 Slide 58 . product. The COCOMO model takes project. 7th edition. personnel and hardware attributes into account when predicting effort required. ©Ian Sommerville 2004 Software Engineering. The time to complete a project is not proportional to the number of people working on the project.

Sign up to vote on this title
UsefulNot useful