Advanced material requirements planning in microelectronics manufacturing

• US 5,943,484 A
• Filed: 09/26/1997
• Issued: 08/24/1999
• Est. Priority Date: 09/26/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yen, Chih-Kuan, Orzell, Robert A., Milne, Robert J.
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): JONES, HUGH M
• Application Number:
  US08/938,130
• Time in Patent Office: 697 Days
• Field of Search: 395/500, 705/28, 364/705, 364/468.13, 364/468.14
• US Class Current: 700/100
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: Y02P 90/80 Management or planning

Abstract

A computer implemented decision support tool serves as an Advanced Material Requirements Planning (AMRP) solver to generate a match between existing assets and demands across multiple manufacturing facilities within the boundaries established by the manufacturing specifications and process flows and business policies to determine what (and when) is needed to start internally or purchase externally to meet all customer demands of current interest. The matching must take into account manufacturing or production specifications and business guidelines. To accomplish the task of deciding what to do when to meet customer demand, the AMRP solver explodes demands into a build plan for purchased and manufacturing orders for end items as well as components and raw materials necessary to produce those end items. The AMRP solver combines traditional MRP decision technology with linear programming (LP) decision technology to provide both speed and intelligence in the matching process.

Claims

1. A computer implemented decision support method which intelligently matches assets with demand to create a material requirements planning (MRP) match to determine what work needs to be accomplished to meet demand comprising the steps of:

accessing a bill of materials in computer memory to identify parts by part numbers of a product to be manufactured;

classifying each part by part number as MRP, Linear Programming (LP) MRP, or non-critical, LPMRP parts being identified from binning data and data related to permissible substitutions and build options as those parts which may be substituted for or by another part number that does not result as output from the same binning process, non-critical parts being identified by a planner using the method as those which can be run with MRP as can components and sub-components of the non-critical parts, and MRP parts being those not classified as LPMRP parts or non-critical parts;

accessing a build options file, a bill of materials file, binning data and permissible substitutions data in computer memory and processing LPMRP parts using a series of integrated LP and MRP methods;

processing MRP parts and non-critical parts using a heuristic method; and

consolidating output from the processing steps using linear programming and heuristic methods and providing a single output file.

2. The computer implemented decision support method recited in claim 1 wherein the step of classifying each part by part number comprises the steps of:

identifying an initial set of non-critical parts;

identifying all non-critical parts by exploding through the accessed bill of material;

identifying an initial set of LPMRP parts by examining situations with alternate processes and/or substitutions outside of binned processes;

identifying a complete set of LPMRP parts by iteratively exploding down and imploding up through the accessed bill of material; and

identifying as MRP parts all parts not previously identified as non-critical parts or LPMRP parts.

3. The computer implemented decision support method recited in claim 1 further comprising the step of transforming a multiple process per part number representation into a single process per part number representation.

4. The computer implemented decision support method recited in claim 1 further comprising the step of determining low-level codes such that each part number has a same low-level code as all parts for which it may substitute, and each part number resulting as output of a binning process has a same low-level code as all other part numbers output from that binning process.

5. The computer implemented decision support method recited in claim 4 wherein the processing of LPMRP parts component comprises the step of running an integrated LP and MRP process at each level of the bill of material.

6. The computer implemented decision support method recited in claim 5 wherein the LPMRP processing further comprises the steps of:

determining whether an LP method is required at each low level code;

invoking an LP method at each required low level code;

taking an output of the LP run and feeding its resulting substitutions (and implicitly alternate processes) into an MRP method for that low level code; and

invoking an MRP method which considers LP substitutions as demands and receipts.

7. The computer implemented decision support method recited in claim 1 wherein the processing of any non-critical parts comprises the steps of:

gathering dependent demand for non-critical parts stemming from the MRP and LPMRP methods; and

executing an MRP run for the non-critical part numbers.