Title

Materials Requirement Planning With A Microcomputer

Comments

Authors: contact us about adding a copy of your work at STARS@ucf.edu

Abbreviated Journal Title

Comput. Ind. Eng.

Keywords

Psychology; Multidisciplinary

Abstract

Materials requirement planning (MRP) has been implemented on equipment ranging from early EAM punchcard equipment to the largest mainframe computers. The availability of relatively capable microcomputers raises the possibility of MRP systems for small operations. A major obstacle is that the concepts underlying most MRP systems are such that both time and memory capability become critical. A system of three programs has been developed using a different computational concept. The “Gozinto” method is used for structuring bill of materials data, and matrix multiplication is used for data manipulation and report generation. The system is reasonably fast, and uses disc memory to alleviate memory constraints. Written in UCSD Pascal, it is usable in any microcomputer supporting that language that has two disc drives and a minimum of 56K of memory. The first program is the bill of materials processor, which receives bill of materials data input from the CRT and stores it on disc using any firm's part coding system. The bill of materials file is then reindexed to develop a triangular matrix. This matrix is then inverted to arrange the bill of materials data in “indented” form. The second “planning” program receives input from the CRT to develop a data base containing master schedule, inventory, order, lot size and other data. The matrix multiplication of the data base and the indented bill of materials is then done on a disc-to-disc basis, obviating the need for large working memory. This second program also provides for revision of the data base as the time periods progress. Results are printed by a third program that essentially despools the matrix products. Limited testing using data from a cooperating firm indicates usefulness with a minimal computer configuration. Extension to larger firms seems feasible with modest memory expansion.

Journal Title

Computers & Industrial Engineering

Volume

9

Issue/Number

230

Publication Date

1-1-1985

Document Type

Article

Language

English

First Page

225

Last Page

142

WOS Identifier

WOS:A1985AQZ8900003

ISSN

0360-8352

Share

COinS