2Software metrics refers to a broad range of measurements for computer software. Measurement can be applied to the software process with the intent of improving it on a continuous basis.Measurement can be used throughout a software project to assist in estimation, quality control, productivity assessment, and project control.Measurement can be used by software engineers to help assess the quality of technical work products and to assist in tactical decision making as a project proceeds.
3Why do we Measure?To characterizeTo evaluateTo predictTo improve
4Measures, Metrics, and Indicators A measure provides a quantitative indication of the extent, amount, dimension, capacity, or size of some attribute of a product or process.Metrics is a quantitative measure of the degree to which a system, component, or process possesses a a given attribute.
5Measures, Metrics, and Indicators An indicator is a metric or combination of metrics that provide insight into the software process, a software project, or the product itself. An indicator provides insight that enables the project manager or software engineers to adjust the process, the project, or the process to make things better.
6Metrics in the Process and Project Domains Process indicators enable a software engineering organization to gain insight into the efficacy of an existing process (I.e., the paradigm, software engineering tasks, work products, and milestones).They enable managers and practitioners to assess what works and what doesn’t.
7Metrics in the Process and Project Domains Project indicators enable a software project manager toassess the status of an ongoing projecttrack potential risksUncover problem areas before they go “critical”Adjust work flow or tasks, andEvaluate the project team’s ability to control quality of software work products
84.2.1 Process Metrics and Software Process Improvement Fig 4.1We measure the efficacy of a software process indirectly; we derive a set of metrics based on the outcomes that can be derived from the process.
9Process Metrics and Software Process Improvement A software metrics etiquette:Use common sense an organizational sensitivity when interpreting metrics dataProvide regular feedback to the individuals and teams who collect measures and metricsDon’t use metrics to appraise individualsWork with practitioners and teams to set clear goals and metrics that will be used to achieve themCont..
10Process Metrics and Software Process Improvement A software metrics etiquette (cont.):Never use metrics to threaten individuals or teamsMetrics data that indicate a problem area should not be considered “negative.” These data are merely an indicator for process improvement.Don’t obsess on a single metric to the exclusion of other important metrics.
11Process Metrics and Software Process Improvement A more rigorous approach: statistical software process improvement (SSPI):All errors and defects are categorized by origin (flaw in spec, flaw in logic, nonconformance to standards).The cost to correct each error and defect is recorded.The number of errors and defects in each category is counted and ranked in descending order. Cont..
12Process Metrics and Software Process Improvement SPPI (cont.):4. The overall cost of errors and defects in each category is computed.5. Resultant data are analyzed to uncover the categories that result in the highest cost to the organization.6. Plans are developed to modify the process with the intent of eliminating (or reducing the frequency of) the class of errors and defects that is most costly.Fig 4.2 and Fig 4.3
134.2.2 Project MetricsProject metrics are used by a project manager and a software team to adapt project work flow and technical activities.Occurred during:estimation monitor and control progress.production rates: pages of documentation, review hours, function points, and delivered source lines.errorstechnical metrics quality
14Project MetricsThe first application of project metrics on most software projects occurs during estimation. Metrics collected from past projects are used as a basis from which effort and time estimates are made from current software work.Production rates are measured.
15Project Metrics The intent of project metrics are two folds: to minimize the development schedule by making the adjustments necessary to avoid delays and mitigate potential problems.to assess product quality on an ongoing basis and, when necessary, modify the technical approach to improve quality.
16Project MetricsAnother model of project metrics suggests that every project should measure:Inputs – measures of the resources required to do the workOutputs – measures of the deliverables or work products created during the software engineering processResults – measures that indicate the effectiveness of the deliverables
17Software MeasurementDirect measures of SE process include cost and effort. Direct measures of product include LOC produced, execution speed, memory size, and defects reported over some set period of time.Indirect measures of product include functionality, quality, complexity, efficiency, reliability, maintainability, and many other “-abilities”
184.3.1 Size-oriented Metrics Derived by normalizing quality and/or productivity measures by considering the size of the software that has been produced.Fig 4.4For example: choose LOC as normalization value.
19Size-oriented Metrics Then we can develop a set of simple size-oriented metrics:Errors per KLOCDefects per KLOC$ per LOCPage of documentation per KLOCAnd other interesting metrics can be computed:Errors per person-month, LOC per person-month, $ per page of documentation.
204.3.2 Function-Oriented Metrics Use a measure of the functionality delivered by the application as a normalization value.Functionality can not be measured directly, it must be derived indirectly using other direct measures.A measure called the function point.
21Function-Oriented Metrics Function points are derived using an empirical relationship based on countable (direct) measures of software's information domain and assessments of software complexity.Function points are computed by completing the table shown in Fig 4.5.
27Typical Function-Oriented Metrics errors per FPdefects per FP$ per FPpages of documentation per FPFP per person-month
284.4.3 Extended Function Point Metrics Function point was inadequate for many engineering and embedded systems.A function point extension called feature points, is a superset of the function point measure that can be applied to systems and engineering software applications.Accommodate applications in which algorithmic complexity is high.
29Extended Function Point Metrics The feature point metric counts a new software characteristic – algorithms.Another function point extension – developed by Boeing integrate data dimension of software with functional and control dimensions. “3D function point”.“Counted, quantified, and transformed”
30Extended Function Point Metrics To compute 3D function points, use this relationship:Index = I + O + Q + F + E + T + RWhere each complexity weighted value is computed using:Complexity weighted value = NilWil+NiaWia+NihWihWhere Nil, Nia, Nih represent the number of occurrences of element I for each complexity; and Wil, Wia, and Wih are the corresponding weights.
31Extended Function Point Metrics Function points, feature points, and 3D point represent the same thing – “functionality” or “utility” delivered by software.
324.4 Reconciling Different Metrics Approaches Attempt to relate FP and LOC measures. Table in page 94
334.5 Metrics for Software Quality Must use technical measures to evaluate quality in objective, rather than subjective ways.Must evaluate quality as the project progresses.The primary thrust is to measure errors and defects metrics provide indication of the effectiveness software quality assurance and control activities.
34Measuring Quality Correctness: defects per KLOC Maintainability: the ease that a program can be corrected, adapted, and enhanced. Time/cost.Time-oriented metrics: Mean-time-to-change (MTTC)Cost-oriented metrics: Spoilage – cost to correct defects encountered.
35Measuring Quality Integrity: ability to withstand attacks Threat: the probability that an attack of a specific type will occur within a given time.Security: the probability that the attack of a specific type will be repelled.Integrity = sum [(1 – threat)x(1 – security)]
36Measuring QualityUsability: attempt to quantify “user-friendliness” in terms of four characteristics:The physical/intellectual skill to learn the systemThe time required to become moderately efficient in the use of the systemThe net increase of productivityA subjective assessment of user attitude toward the system (e.g., use of questionnaire).
37Defect Removal Efficiency A quality metric that provides benefit at both the project and process level.DRE is a measure of filtering ability of quality assurance and control activities as they applied throughout all process framework activities.
38Defect Removal Efficiency DRE = (errors) / (errors + defects)whereerrors = problems found before releasedefects = problems found after releaseThe ideal value for DRE is 1 no defects found.
39Defect Removal Efficiency DRE is defined as:DRE = E/(E + D)Where E is the number of errors found before delivery of S/W to the end-userAnd D is the number of defects found after deliveryThe ideal value for DRE is 1 no defects found.
404.6 Integrating Metrics Within the Software Process Arguments for Software Metrics:Why is it so important to measure the process of software engineering and the product (software) that it produces?
414.7 Managing Variation: Statistical Process Control How can we compare a variety of different projects?Use of Control Chart: to determine whether the dispersion (variability) and “location” (moving average) of process metrics are stable or unstable.The moving average control chartThe individual control chartFig. 4.8 Page102
42Moving Range (mR) Control Chart Calculate the moving ranges (mR)Calculate the mean of the moving rangesMultiply the mean by upper control limit (UCL)Fig. 4.8 4.9Are all moving range values inside the UCL?If “yes” stable
43Individual Chart Control Plot individual metrics values as shown in Fig 4.8Compute the average value, AmMultiply the mean of the mR value by and add Am in (2) plot the upper natural process limit (UNPL)Multiply the mean of the mR value by and subtract Am in (2) plot the lower natural process limit (LNPL)Compute the SD as (UNPL – Am)/3. Plot lines one and two SD above and below Am.
44Individual Chart Control Zone rules: If any of the following conditions is true, the metrics data is out of control:A single metrics value lies outside the UNPLTwo out of three successive metrics values lie more than two SD away from AmFour out of five successive metric values lie more than one SD away from AmEight consecutive metrics values lie on one side of Am.
454.8 Metrics for Small Organizations “Keep it simple”:TimeEffortErrorsDefects
46Homework #2 Problem# 4.9, 4.11, 4.13, 4.17, and 4.18 Due Mon 15 July 2002