Barry W. Boehm

from Wikipedia, the free encyclopedia
Barry Boehm 2006

Barry W. Boehm (born May 16, 1935 in Santa Monica , California ) is an American software engineer , who was known for his cost considerations of large software projects . He saw software as the primary cost factor for future information technologies . So he invented the COCOMO software cost calculation, the risk and cost-reducing procedure - spiral model and u. a. an extended Delphi method ( wideband delphi ). He has worked as a consultant and researcher on many major military and civil projects.

Boehm earned his bachelor's degree in mathematics from Harvard University in 1957 , and in 1964 he received his PhD in mathematics from the University of California, Los Angeles . After that he worked for many years as a teacher and researcher. He later served as a consultant for RAND Corporation , TRW and DARPA . Today he is Professor of Software Engineering, Computer Science, at the University of Southern California (USC).

Software cost estimation using the COCOMO method

Barry W. Boehm's 1981 book Software Engineering Economics described, among other things, his Constructive Cost Model ( COCOMO ).

COCOMO (81)

COCOMO is an algorithmic model for cost and effort estimation for the development of computer programs. The effort (measured in Staff Month (152 working hours)) is calculated depending on the software size, which is measured in Delivered Source Instructions (DSI), using statistical calculation methods based on data collected from past projects. COCOMO divides the projects into three groups according to their size. The effort is calculated in three degrees of detail, with the simplest model providing a rather rough estimate:

  • = constant scaling coefficient for the effort
  • = This exponent represents the savings through increased production (economies of scale); especially the savings by avoiding activities that do not add value (e.g. idle times, delays, communication overhead, etc.)

The other calculation models include various cost factors (e.g. experience of the programmer, quality requirements of the software, etc.) and divide the project into various sub-components and the project into various phases. COCOMO thus achieves a relatively high degree of accuracy.

ADA COCOMO

ADA COCOMO is a further development of COCOMO 81 - which is heavily influenced by batch processing and the waterfall process model - for adaptation to the TRW implementation of the ADA process model. This model includes risk management, architecture skeletons, incremental implementation and testing, and uniform software metrics. The main focus of the model is on reducing communication overhead and avoiding late revisions due to unstable requirements.

COCOMO II

Like its two predecessors, COCOMO II was also developed by Barry W. Boehm and first published in 1997. The officially known version was published in 2000 with the book Software Cost Estimation with COCOMO II . COCOMO II represents the changes in "modern" software development. As in COCOMO 81, a distinction is again made between three different types of estimation, with the difference, however, that these increasingly relate to the stage of development of the project, but the subdivision according to project size is dispensed with.

The process model spiral model

Based on his extensive experience, Boehm developed a new process model for software development. This spiral model is intended to reduce the risk of failure by repeatedly viewing the phase results in a spiral. The spiral model has influenced many subsequent procedural models.

Wideband Delphi

Boehm refined the Delphi method and called this refined method Wideband Delphi . This made it applicable to a larger area.

Awards

  • 1992 - Award for Excellence (Office of the Secretary of Defense).
  • 1994 - Lifetime Achievement Award from ASQC .
  • 1997 - Distinguished Research Award in Software Engineering from ACM .
  • 2005 - Mellon Award for Excellence in Mentoring from USC.

He is a fellow of the AIAA , the ACM and the IEEE . He is also a member of the National Academy of Engineering .

literature

  • Barry W. Boehm: Software Engineering Economics , Englewood Cliffs, NJ, Prentice-Hall, 1981. ISBN 0-13-822122-7

Web links

Commons : Barry W. Boehm  - Collection of images, videos and audio files

Individual evidence

  1. Counting Lines of Code (LOC) or Source Lines of Code (SLOC) is problematic and has often been criticized. A corresponding summary of the problem is currently only available in English