System design is an NP-complete problem

William L. Chapman; Jerzy Rozenblit; A. Terry Bahill

doi:10.1002/sys.1018

System design is an NP-complete problem

William L. Chapman, Jerzy Rozenblit, A. Terry Bahill

Research output: Contribution to journal › Article › peer-review

33 Scopus citations

Abstract

The system design process translates the customers' needs into a buildable system design. It requires selecting subsystems from an allowable set and matching the interfaces between them. Designs that meet the top-level input and output requirements are tested to see how well they meet the system's performance and cost goals. This paper proves that the System Design Problem is NP-complete by reduction from the Knapsack Problem, which is known to be NP-complete. The implication of this proof is that designing optimal systems with deterministic, polynomial time procedures is not possible. This is the primary reason why engineers do not try to produce optimal systems: They merely produce designs that are good enough.

Original language	English (US)
Pages (from-to)	222-229
Number of pages	8
Journal	Systems Engineering
Volume	4
Issue number	3
DOIs	https://doi.org/10.1002/sys.1018
State	Published - 2001

ASJC Scopus subject areas

Hardware and Architecture
Computer Networks and Communications

Access to Document

10.1002/sys.1018

Cite this

@article{66b41c2cae6647959e663147f077fc91,

title = "System design is an NP-complete problem",

abstract = "The system design process translates the customers' needs into a buildable system design. It requires selecting subsystems from an allowable set and matching the interfaces between them. Designs that meet the top-level input and output requirements are tested to see how well they meet the system's performance and cost goals. This paper proves that the System Design Problem is NP-complete by reduction from the Knapsack Problem, which is known to be NP-complete. The implication of this proof is that designing optimal systems with deterministic, polynomial time procedures is not possible. This is the primary reason why engineers do not try to produce optimal systems: They merely produce designs that are good enough.",

author = "Chapman, {William L.} and Jerzy Rozenblit and Bahill, {A. Terry}",

year = "2001",

doi = "10.1002/sys.1018",

language = "English (US)",

volume = "4",

pages = "222--229",

journal = "Systems Engineering",

issn = "1098-1241",

publisher = "John Wiley & Sons Inc.",

number = "3",

}

TY - JOUR

T1 - System design is an NP-complete problem

AU - Chapman, William L.

AU - Rozenblit, Jerzy

AU - Bahill, A. Terry

PY - 2001

Y1 - 2001

N2 - The system design process translates the customers' needs into a buildable system design. It requires selecting subsystems from an allowable set and matching the interfaces between them. Designs that meet the top-level input and output requirements are tested to see how well they meet the system's performance and cost goals. This paper proves that the System Design Problem is NP-complete by reduction from the Knapsack Problem, which is known to be NP-complete. The implication of this proof is that designing optimal systems with deterministic, polynomial time procedures is not possible. This is the primary reason why engineers do not try to produce optimal systems: They merely produce designs that are good enough.

AB - The system design process translates the customers' needs into a buildable system design. It requires selecting subsystems from an allowable set and matching the interfaces between them. Designs that meet the top-level input and output requirements are tested to see how well they meet the system's performance and cost goals. This paper proves that the System Design Problem is NP-complete by reduction from the Knapsack Problem, which is known to be NP-complete. The implication of this proof is that designing optimal systems with deterministic, polynomial time procedures is not possible. This is the primary reason why engineers do not try to produce optimal systems: They merely produce designs that are good enough.

UR - http://www.scopus.com/inward/record.url?scp=1642295856&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1642295856&partnerID=8YFLogxK

U2 - 10.1002/sys.1018

DO - 10.1002/sys.1018

M3 - Article

AN - SCOPUS:1642295856

SN - 1098-1241

VL - 4

SP - 222

EP - 229

JO - Systems Engineering

JF - Systems Engineering

IS - 3

ER -

System design is an NP-complete problem

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this