- #Cocomo model for un manned aerial vehicle software#
- #Cocomo model for un manned aerial vehicle professional#
Sensor Control and Signal Processing (SCP) 2. SRDR dataset was segmented into 14 productivity types to increase the accuracy of estimating cost and schedule 1.
#Cocomo model for un manned aerial vehicle software#
Office of Program Accountability and Risk Management 2ģ OUTLINE Research Method Data Demographics Software Productivity Benchmarks Effort and Schedule Estimation Models Conclusion Backup Office of Program Accountability and Risk Management 3ĥ Instrumentation Questionnaire: Software Resource Data Report (SRDR) (DD Form 2630) Source: Defense Cost Analysis Resource Center (DCARC) website: Content: Allows for the collection of project context, t company information, requirements, product size, effort, schedule, and quality Office of Program Accountability and Risk Management 5Ħ Data Collection and Validation Initial Dataset 800 completed software projects were collected from DCARC Of the 800 projects, 345 were fully reviewed using GAO Best Practices Of the 345 reviewed, 141 were excluded based on the following limitations: Inadequate information on reused and modified code Projects cancelled or terminated before delivery Missing/Inaccurate effort and schedule data Same duration (start and end dates) across software projects/components Missing Adaptation and Adjustment Factors (DM, CM, IM) Duplicate records or submissions Estimates At Completion vice Actual Data Final Dataset 204 of 345 projects included in the analysis as these passed quality inspection Office of Program Accountability and Risk Management 6ħ Data Normalization and Analysis Workflow Data was normalized to account for cost and sizing units, mission or application, technology maturity, and content so they are consistent for comparisons (source: GAO) Segment Data by Operating Environment Segment Data by Productivity it Type Convert Size to Logical Count Normalize to Equivalent Size Analyze with Descriptive Statistics Select ect Best Model Form Office of Program Accountability and Risk Management 7Ĩ Segment Data by Operating Environment (OE) Operating Environment Acronym Examples Ground Site Ground Vehicle Maritime Vessel Aerial Vehicle Fixed GSF Command Post, Ground Operations Center, Ground Terminal, Test Faculties Mobile GSM Intelligence gathering stations mounted on vehicles, Mobile missile launcher Manned GVM Tanks, Howitzers, Personnel carrier Unmanned GVU Robots Manned MVM Aircraft carriers, destroyers, supply ships, submarines Unmanned MVU Mine hunting systems, Towed sonar array Manned AMV Fixed-wing aircraft, Helicopters Unmanned AVU Remotely piloted air vehicles Ordinance Vehicle Unmanned OVU Air-to-air missiles, Air-to-ground missiles, Smart bombs, Strategic missiles Space Vehicle Manned SVM Passenger vehicle, Cargo vehicle, Space station Unmanned SVU Orbiting satellites (weather, communications), Exploratory space vehicles Office of Program Accountability and Risk Management 8ĩ Segment Data by Productivity Type (PT) Different productivities have been observed for different software application types. Analysis results will be discussed in this presentation. Over 200 actual software projects from DoD s Software Resource Data Reports (SRDRs) were fully inspected and analyzed to produce a comprehensive set of Cost Estimation Relationships, Schedule Estimation Relationships, and Software Productivity Benchmarks. Consideration is also given to the operating environment it operates within. Productivity types are groups of application domains that are environment independent, technology driven, and are characterized by 13 COCOMO product attributes.
#Cocomo model for un manned aerial vehicle professional#
1 Domain-Driven Software Cost, Schedule, and Phase Distribution Models: Using Software Resource Data Reports Wilson Rosa (DHS) Barry Boehm (USC) Brad Clark (SEI-CMU) Ray Madachy (NPS) Joseph P Dean (AFCAA) 2013 ICEAA Professional Development & Training Workshop J(New Orleans, LA)Ģ Introduction Instead of developing software cost and schedule estimation models with many parameters, this paper describes an analysis approach based on grouping similar software applications together called Productivity Types.
0 kommentar(er)
