제1차 S/W Management 연구회 ( 경영정보학회 ) 2004년 4월17일 정호원교수 ( 고려대학교경영학과 ) SPICE Research International Coordinator SEI Authorized CMMI Instructor IPRC Charter Research Member hwjung@korea.ac.kr http://biz.korea.ac.kr/~misjung (T) 3290-1938 1
- 경쟁력향상 - 고객만족 2
소프트웨어관련문제점 -1 전세계의대형프로젝트의 1% 만이목표달성 일정 계획에비해평균 1년지연 비용 계획에비해평균 2배소용 품질 수많은error 고객불만족 Möller, K.-H. and Paulish, D.J., 1993, Software Metrics: A Practitioner s Guide to improve Product Development (New York: Chapman & Hall Computing). 에러한개수정평균비용 : $7,000 at Ericsson Ohlsson, N. and Albert, H, 1996, Predicting fault-prone software modules in telephone switches. IEEE Trans. on Software Engineering, 22, (12), 886-894. 3
소프트웨어관련문제점 -2 미국의 business 부분의소프트웨어년간투자 2,500 억달러 소프트웨어에러년간비용 1250 억달러 Windows 95 defects: 10,000 1995년조사통계 폐기된프로젝트 : 31.1% 비용과다와일정지연프로젝트 : 52.7% (189% 의비용추가 ) 성공프로젝트 : 16.2% A 1996 study by the Standish Group 4
프로세스관련문제점해결방법 문제해결방안 기술적인해결방안 New programming language, new methodology, etc. 프로세스개선 능력향상 (capability improvement) 프로세스표준의역할 5
능력 (capability) 이란? 5 단계 ( 고수의세계 ) 4 단계 ( 선형대수 : AX=b) 3 단계 ( 연립방정식 : 2x+ y =3; x+2y = 3) 2 단계 ( 미지수구하기 : 2x + 3 = 9) 1 단계 (4 칙연산 : 더하기, 빼기, 곱하기, 나누기 ) 0 단계 (?) 능력에는단계가있다. 즉, 사칙연산을모르면미지수문제의답을구할수없다 6
심사와능력의관계 심사를통하여현재의능력을파악 - 각프로세스의강점과약점파악 우선순위를고려한프로세스개선 현재의능력을모르고는개선 ( 지도 ) 은불가능 7
ISO/IEC 15504(SPICE) CMMs 8
Member only: http://wg10.intranets.com/ Subscription: http://www.isospice.com/ Research: http://www.cis.strath.ac.uk/research/spice/ KSPICE: http://www.kspice.co.kr/ ( 한국 ) 한국의 ISO/IEC 15504 심사원시험합격자 : 약 600 명 9
ISO/IEC 15504 현황 ISO/IEC FDIS 15504: Information Technology Process Assessment Part 1: Concepts and Vocabulary ( 지침 ) Part 2: Performing an Assessment ( 표준 ) Part 3: Guidance on performing an assessment ( 지침 ) Part 4: Guidance on use for process improvement and process capability determination ( 지침 ) Part 5: An exemplar process assessment model ( 지침 ) 10
Two-dimensional architecture 3 Categories 10 Groups 48 processes 11
Process dimension 3 Categories 10 Groups 48 processes 12
Capability level and process attributes Capability level Level 0 (Incomplete) Level 1 (Performed) Level 2 (Established) Level 3 (Managed) Level 4 (Predictable) Level 5 (Optimizing) Process attributes PA1.1: Process Performance PA2.1: Performance Management PA2.2: Work Product Management PA3.1: Process Definition PA3.2: Process Deployment PA4.1: Process Measurement PA4.2: Process Control PA5.1: Process Innovation PA5.2: Process Optimization 13
14
Capability and maturity Models Reference (Assessment) Models SW-CMM (Software CMM) Capability Maturity Model for Software, Version 1.1 (CMU/SEI-93-TR-024) Key Practices of the Capability Maturity Model, Version 1.1 (CMU/SEI-93-TR-025) SA-CMM (Software Acquisition CMM) SE-CMM (Systems Engineering CMM) P-CMM (People Capability Maturity Model IPD-CMM (Integrated Product Development CMM) CMMI (CMM Integration) CMMI for Systems Engineering/Software Engineering/Integrated Product and Process Development/Supplier Sourcing, Version 1.1, Continuous Representation (CMMI-SE/SW/IPPD/SS, V1.1) (Continuous: CMU/SEI-2002-TR-011) and Staged Representation (CMU/SEI-2002-TR-012) Appraisal (Assessment) Methods SPA (Software Process Assessment) Clarity Testing Interim Profile CBA IPI (CMM-Based Appraisals for Internal Process Improvement) CMM-Based Appraisal for Internal Process Improvement (CBA IPI) Lead Assessor s Guide, V1.1 (CMU/SEI-96-HB-003) CMM-Based Appraisal for Internal Process Improvement (CBA IPI): Method Description (CMU/SEI-96-TR-007, ADA307934) SCE (Software Capability Evaluation) CMMI SCAMPI (Standard CMMI Appraisal Method for Process Improvement) Standard CMMI Appraisal Method for Process Improvement (SCAMPI), Version 1.1: Method Definition Document (CMU/SEI-2001-HB-001) Others 15
SW-CMM 의폐기 SEI 는 SW-CMM 은 2003 년말폐기 10 년전에만들어진기준 어떠한기관도 SW-CMM 을 SEI 을대신해보급불가 더이상심사원교육없음 폐기이유 SW-CMM 의내용이잘못된것이아님 SW-CMM 의내용이미국방성의목표와다름 CBA-IPI 선임심사원자격 앞으로 2 년간유효 (SEI 는더이상계약을않음 ) 이후는더이상선임심사원이란용어를사용않음 SEI 는 CBA-IPI 심사결과는 SEI 의 PAIS DB 에서는수용 연구목적으로만사용 16
- Capability Maturity Model Integration - SEI: http://www.sei.cmu.edu/ (CMMI) CMMI: http://www.sei.cmu.edu/cmmi/ (CMMI) SEI는 CMMI가 ISO/IEC 15504의 compatible 하다고선언 약 3년간모형변경없음 17
Process 표준의문제점 Systems Engineering Software Engineering 시스템공학과소프트웨어공학은전통적으로다른분야로여겨져옴시스템에서소프트웨어의중요성증대 예 : 시스템에서소프트웨어가차지하는비중 (%) B-2 -- 65% F-22 -- 80% 미국방성은시스템공학과소프트웨어공학을통합한평가모형을개발 18
시스템공학과소프트웨어공학통합 Systems Engineering CMMI Software Engineering 시스템공학과소프트웨어공학통합 두공학분의개선을동시에추구 19
Two Representations Continuous Staged Process Area Capability 0 1 2 3 4 5 ML5 ML4 ML3 ML2 ML 1 PA1 PA2 PAn...for (1) a single process area or (2) a set of process areas...for (1) an established set of process areas across an organization 20
Process Area Capability Profile 21
Continuous representation SE/SW/IPPD/SS, Version 1.1 (25 Process Areas) 22
Staged representation SE/SW/IPPD/SS, Version 1.1 (25 Process Areas) 23
24
Product type Software systems Software + hardware 통신기기 컴퓨터주변기기 무기 system The Best? Application domain Finance (excluding banking) Insurance Banking Petroleum Automotive Public Utilities (Gas, Water, Electricity) Aerospace Telecommunications Public Administration Consumer Goods Retail Distribution/Logistics Defense Information Technology / Software Health and Pharmaceutical Leisure and Tourism Manufacturing Construction Travel Media (Television, Radio) Education 25
The Best? 우리의선택에서고려해야할사항 Market requirements 비용 Appraisal Document purchase Training Maintainability of standards 비교가능한사례들 26
27
프로세스개선의목표 Process improvement benefits fall into one or more general categories Improved schedule and budget predictability Improved cycle time Improved quality (as measured by defects) Increased productivity Increased customer satisfaction Improved employee morale Increased return on investment Decreased cost of quality called external attributes (Performance criteria) 28
이론적배경 Factor A Factor B Process improvement Performance Factor C Factor D Is there an association between process improvement and performance? 29
성과변수의측정방법 Considering process context Examples of process context Organization size Application domains ISO 9000 certification 30
Continuous model vs. Staged model Continuous models ISO/IEC 15504 CMMI Continuous Representation Staged models SW-CMM CMMI Staged Representation 31
Functional form Continuous model: Correlation[CL(x), Performance] Staged model: Correlation[CL( X + L + X ), Performance] 1 k Staged model에서는어떤 process가 improve 되어 performance가 improve되었는지알수없다. 따라서, -SW-CMM 의 CBI-IPA method, CMMI Staged representation 을 사용해서는 level 향상이성과와관련이있다는것을실증적으로보여주기에는많은시간이걸림 32
33
통용되는상식 어떠한기술이나 model 도 70% 이상의시장을점유하지어렵다. 5 년이상시장을지배하기어렵다. 다른기술이나 model 을사용하는사람이나기업도적이아니다. 34
George Box 35
36
Assessor Training SPICE Quality Academy KSPICE and KASPA: Lead assessor CMMI Introduction to CMMI Intermediate (SEI) SCAMPI (SEI) Instructor (SEI) 37