US20060069597A1 - Pegging resources to shipments in a pre-existing production plan - Google Patents

Pegging resources to shipments in a pre-existing production plan Download PDF

Info

Publication number
US20060069597A1
US20060069597A1 US10/953,181 US95318104A US2006069597A1 US 20060069597 A1 US20060069597 A1 US 20060069597A1 US 95318104 A US95318104 A US 95318104A US 2006069597 A1 US2006069597 A1 US 2006069597A1
Authority
US
United States
Prior art keywords
tr
production plan
pegging
resource
steps
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US10/953,181
Inventor
Robert Wittrock
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
International Business Machines Corp
Original Assignee
International Business Machines Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by International Business Machines Corp filed Critical International Business Machines Corp
Priority to US10/953,181 priority Critical patent/US20060069597A1/en
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION reassignment INTERNATIONAL BUSINESS MACHINES CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: WITTROCK, ROBERT JAMES
Publication of US20060069597A1 publication Critical patent/US20060069597A1/en
Application status is Abandoned legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management, e.g. organising, planning, scheduling or allocating time, human or machine resources; Enterprise planning; Organisational models
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management, e.g. organising, planning, scheduling or allocating time, human or machine resources; Enterprise planning; Organisational models
    • G06Q10/063Operations research or analysis
    • G06Q10/0631Resource planning, allocation or scheduling for a business operation
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management, e.g. organising, planning, scheduling or allocating time, human or machine resources; Enterprise planning; Organisational models
    • G06Q10/063Operations research or analysis
    • G06Q10/0631Resource planning, allocation or scheduling for a business operation
    • G06Q10/06312Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling

Abstract

In software systems for resource-constrained production planning (RCPP), it is often beneficial to determine a detailed association, or “pegging”, between the resources and the shipments that are enabled by those resources. The method implemented in the software systems is called “Pegging By Explosion” (PBE), for constructing a pegging of a production plan for an RCPP problem. The Pegging By Explosion technique takes a feasible production plan and reconstructs it, using its own specialized incremental allocation technique. A Pegging By Monitoring technique is applied to this reconstruction in order to produce the pegging. The reconstruction is accomplished by a specialized “explosion” technique related to the explosion technique of Material Requirements Planning (MRP).

Description

    BACKGROUND OF THE INVENTION
  • 1. Field of the Invention
  • The present invention generally relates to manufacturing resource planing, and more particularly to software systems for resource-constrained production planning.
  • 2. Background Description
  • Software systems for resource-constrained production planning (RCPP) take as input data a description of a production planning problem. This includes a listing of the basic data objects of the problem; i.e., capacity resources, raw material resources, subassemblies, final products, and demands for the products, as well as a multi-level Bill-Of-Resources network that defines how the products can be produced from subassemblies, raw materials and capacities. It also includes various attributes on these objects such as supply volumes, demand volumes, production usage rates, and economic data or priorities. In addition, the problem data is defined with respect to a set of discrete time periods.
  • Given these inputs, an RCPP system computes a production plan that consists of two portions: (1) a production portion of plan that specifies of how much of each product is to be produced in each time period, and (2) a shipment portion of the plan that specifies how much of each demand is to be fulfilled in each time period. (The fulfillment of a demand in a period may be called a “shipment” to that demand.) The production plan computed by the RCPP system is normally required to be feasible, meaning it satisfies the resource constraints that are specified in the input data.
  • Two methods for solving the RCPP problem are described in U.S. Pat. No. 5,548,518 for “Allocation Method for Generating a Production Schedule” and U.S. Pat. No. 5,630,070 for “Optimization of Manufacturing Resource Planning”, both to Dietrich et al. More specifically, U.S. Pat. No. 5,548,518 discloses an allocation method which, in response to a specified requirement, determines what quantity of a product can be produced with a specified quantity of supply components, allocates a required quantity of supply components for filling a defined partial order and fills a remainder of the requirement at some later time. U.S. Pat. No. 5,630,070 discloses a method for constrained material requirements planning, optimal resource allocation and production planning which optimizes a manufacturing process by designating the amounts of various manufactured products to be produced, which products include both end products as well as subassemblies.
  • Users of RCPP systems frequently desire more information than just the production plan itself. One question that is often asked is, “For any given shipment in the production plan, which resources were utilized specifically to enable that shipment?” Generally, this information is not evident in the production plan itself. The production plan specifies which resources are used and which demands are met, but it does not provide an association between the two. This may be called “the pegging problem” for RCPP; that is, given an RCPP problem and a corresponding feasible production plan, define an association between the individual resources that are utilized in the plan and the individual shipments in the plan. Such an association is called a “pegging” of the production plan. A resource that is associated with a particular shipment is said to be “pegged” to it.
  • Users of RCPP systems have many uses for a pegging of a production plan, including the following:
      • Verifying the validity of the production plan.
      • Verifying the validity of the input data.
      • Identifying which demands are responsible for depleting the supplies of critical (i.e., bottleneck) resources.
      • Identifying opportunities to improve the production plan by increasing the supplies of the resources that were used by the most important demands.
      • Generally deepening the user's understanding of the production plan.
  • One possible method for computing a pegging might be called “Pegging By Monitoring”. This technique would apply when the RCPP system solves the RCPP problem by an “incremental allocation” technique, as disclosed for example in U.S. Pat. No. 5,548,518. An incremental allocation technique for RCPP has the following key properties:
      • It starts with an empty production plan.
      • It proceeds by iterations.
      • Each iteration creates only one new shipment.
      • As the technique progresses, the production volume for any product in any period can be increased, but it can never be decreased.
        The “Pegging By Monitoring” technique would proceed by monitoring the incremental allocation technique; that is, whenever a production volume of a product is increased by the incremental technique, that increase is pegged to the shipment for the current iteration.
  • In some respects, “Pegging By Monitoring” is a very appealing form of pegging. It s a direct reflection of what how the production plan was built up by the incremental allocation technique. But it has the following drawbacks:
      • It can only be applied, if the production plan is being computed by an incremental allocation technique.
      • It must be integrated with the RCPP solver itself; it cannot be applied to a production plan that has already been computed.
    SUMMARY OF THE INVENTION
  • It is therefore an object of the present invention to provide a method for solving the pegging problem in a resource-constrained production planning system which avoids the problems of the “Pegging By Monitoring” approach.
  • According to the invention, there is provided a method for constructing a pegging of a production plan for the RCPP problem which is called herein “Pegging By Explosion” (PBE). The Pegging By Explosion technique takes a feasible production plan and reconstructs it, using its own specialized incremental allocation technique. The Pegging By Monitoring technique (described above) is applied to this reconstruction in order to produce the pegging. The reconstruction is accomplished by a specialized “explosion” technique related to the well-known explosion technique of Material Requirements Planning (MRP). (For details above MRP, see Manufacturing Planning and Control Systems, by T. E. Vollmann, W. L. Berry, and D. C. Whybark, McGraw-Hill, 4th ed., Mar. 1, 1997). A conventional MRP explosion technique would usually not be able to reconstruct an arbitrary production plan to an RCPP problem. In contrast, the explosion technique used by PBE does so, by making use of detailed knowledge of the production plan itself.
  • The advantages of using the Pegging By Explosion technique instead of the Pegging By Monitoring technique are as follows:
      • It can be used regardless of how the production plan was generated.
      • It does not need to be integrated with the RCPP solver.
        Note that a pegging of a production plan is not uniquely determined by the production plan itself. For any given feasible production plan, there will normally be many different possible peggings. In light of this, it is important to construct a pegging that a user is likely to find “plausible”. Specifically, it should be plausible that the resources that are pegged to a particular shipment were precisely those resources that made that shipment possible. The PBE technique of the present invention constructs a pegging that is guaranteed to satisfy this “plausibility” condition.
    BRIEF DESCRIPTION OF THE DRAWINGS
  • The foregoing and other objects, aspects and advantages will be better understood from the following detailed description of a preferred embodiment of the invention with reference to the drawings, in which:
  • FIG. 1 is a data flow diagram illustrating the PBE technique in the context of an RCPP system;
  • FIG. 2 is a data flow diagram similar to FIG. 1 with the addition of a shipment sequence;
  • FIG. 3 is a flow diagram illustrating the logic of the PBE algorithm according to the invention;
  • FIG. 4 is a flow diagram of the procedure “Explode-Triple” used in the algorithm of FIG. 3; and
  • FIG. 5 is a flow diagram of the procedure “Main-Iteration” used in the procedure of FIG. 4.
  • DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS OF THE INVENTION
  • The invention disclosed here is a novel method for constructing a pegging of a production plan for an RCPP problem. The technique is called “Pegging By Explosion” (PBE).
  • The Pegging By Explosion technique takes a feasible production plan and reconstructs it, using its own specialized incremental allocation technique. The Pegging By Monitoring technique (described above) is applied to this reconstruction in order to produce the pegging. The reconstruction is accomplished by a specialized “explosion” technique related to the well-known explosion technique of Material Requirements Planning (MRP). (For details above MRP, see Manufacturing Planning and Control Systems, by T. E. Vollman, W. L. Berry, and D. C. Whybark.) A conventional MRP explosion technique would usually not be able to reconstruct an arbitrary production plan to an RCPP problem. In contrast, the explosion technique used by PBE does so by making use of detailed knowledge of the production plan itself.
  • The data flow for the PBE technique in the context of an RCPP system is illustrated in FIG. 1. As shown in FIG. 1, the RCPP problem data 1 is input to the RCPP solver 2, which produces a production plan 3. The Pegging by Explosion (PBE) technique 4 according to the present invention is applied to the thus produced production plan 3, yielding the pegging 5 of the production plan.
  • The advantages of using the Pegging By Explosion technique instead of the Pegging By Monitoring technique are as follows:
      • It can be used regardless of how the production plan was generated.
      • It does not need to be integrated with the RCPP solver.
        Note that a pegging of a production plan is not uniquely determined by the production plan itself: For any given feasible production plan, there will normally be many different possible peggings. In light of this, it is important to construct a pegging that a user is likely to find “plausible”. Specifically, it should be plausible that the resources that are pegged to a particular shipment were precisely those resources that made that shipment possible. Hereinafter, a precise mathematical condition will be defined that formalizes of the idea of a plausible pegging. The PBE technique constructs a pegging that is guaranteed to satisfy this “plausibility” condition.
  • This is an algorithmic invention. It would be implemented in computer software, normally as an additional capability of an RCPP system, although in principle, it could be implemented as a stand-alone software tool.
  • Before describing the PBE technique, the underlying RCPP problem must be defined.
  • The RCPP Problem
  • An RCPP problem is built up out of several kinds of data objects:
    • Periods: A period represents a distinct interval of time, such as a day, a week, two months, a year, etc. The periods in an RCPP problem need not be of uniform duration. The periods are represented by the integers from I to Nperiods, the number of periods.
    • Resources: These represent capacity resources, raw material resources, subassembly resources, and even the final products themselves.
    • Materials: These are resources that remain available in the next period, if they are not used in a given period.
    • Capacities: These are resources that do not remain available in the next period, if they are not used in a given period. Each resource is either a material or a capacity, but not both.
    • Products: These are the resources that can be produced. They may be either materials or capacities.
    • Operations: An operation represents a means by which a particular product can be produced. Each operation produces exactly one product; each product is produced by one or more operations. If a product is produced by more than one operation, this simply indicates that there is more than one way to produce that product.
    • Demands: Each demand is associated with a resource and represents demand for that resource. Each resource is associated with zero or more demands.
      Let
  • AllPer=The set of all periods={1, 2, . . . , Nperiods}
  • AllRes=The set of all resources
  • AllMat=The set of all materials (Thus, AllMatAllRes.)
  • AllCap=The set of all capacities (Thus, AllCapAllRes.)
  • AllOpn=The set of all operations
  • AllDem=The set of all demands
  • ∀dεAllDem:
  • Let DemRes (d)=The resource demanded by demand d.
  • ∀rεAllRes:
  • Let ResDems (r)={dεAllDem|r=DemRes (d)}
  • ∀jεAllOpn:
  • Let ProdRes (j)=The resource produced by operation j.
  • ∀rεAllRes:
  • Let ProdOpns (r)={jεAllOpn|r=ProdRes (j)}
      • [These are the “producing operations” for resource r.]
      • [Resource r is a product, if and only if ProdOpns (r)≠Ø.]
        ∀jεAllOpn:
  • Let BOR (j)=The set of all resources that must be consumed by operation j in order to produce ProdRes (j). This is the “Bill-Of-Resources” of operation j.
  • ∀rεAllRes:
  • Let ConsOpns (r)={jεAllOpn|rεBOR (j)}
      • [These are the “consuming operations” of resource r.]
  • Given any resource r and any period tr, there is an associated nonnegative quantity, SupplyVol (r, tr) (“supply volume”), that specifies the amount of resource r that is available as supply in period tr. Given any demand d and any period td, there is an associated nonnegative quantity, DemandVol (d, td) (“demand volume”), that specifies the amount of resource DemRes (d) that is demanded by demand d in period td.
  • Given any operation j, any period tj, any resource rεBOR (j), and any period tr, there is an associated quantity, UsageRate (j, tj, r, tr) (“usage rate”), that specifies the amount of resource r that must be consumed in period tr for each unit of ProdRes (j) that is produced by operation j in period tj. Note that UsageRate (j, tj, r, tr) may be negative. This indicates that, when operation j produces ProdRes (j) in period tj, it consumes a negative amount of resource r in period tr. This can be used to model operations that have two or more outputs: ProdRes (j) is the main product, while resources rεBOR (j) with UsageRate (j, tj, r, tr)<0 are by-products. Many MRP databases contain negative usage rates in their BOR data.
  • With each resource r, there is an associated integer, ResIdx (r) (“resource index”), distinct for each resource. This index defines an ordering for the resources and has the following property:
  • ∀jεAllOpn, tjεAllPer, ∀rεBOR (j), trεAllPer:
  • If UsageRate (j, tj, r, tr)>0, then:
      • ResIdx (ProdRes (j))>ResIdx (r).
        Finally, there would also be economic or priority data associated with the resources, operations, and demands, but this data is not necessary for defining a pegging technique.
  • A solution to an RCPP problem is a production plan. Given any operation, j, and any period, tj, the production plan specifies an associated quantity, ProdVol (j, tj) (“production volume”) that represents the amount of resource ProdRes (j) that is to be produced by operation j in period tj. Given any resource, r, and any period, tr, the production plan specifies an associated quantity, StockVol (r, tr) (“stock volume”) that represents the amount of resource r that is to be carried over from period tr to period tr+1. If r is a capacity, then StockVol (r, tr)=0. For notational convenience, let StockVol (r, 0)=0. Given any resource, r, and any period, tr, the production plan specifies an associated quantity, ConsSupVol (r, tr) (“consumed supply volume”) that represents the amount of the supply of resource r from period tr that is consumed in any period. Given any demand, d and any period td, the production plan specifies an associated quantity, ShipVol (d, td) (“shipment volume”) that represents the quantity of the demand that is to be fulfilled (i.e., “shipped”) in period td. The collection of production volumes of all operations, stock volumes and consumed supply volumes of all resources, and shipment volumes of all demands constitutes the production plan.
  • Given a production plan, the following further quantities may be defined:
    ∀rεAllRes, ∀trεAllPer: Let TotShipVol ( r , tr ) = d ResDems ( r ) ShipVol ( d , tr ) Let TotProdVol ( r , tr ) = j ProdOpns ( r ) ProdVol ( j , tr ) Let ConsVol ( r , tr ) = j ConsOpns ( r ) tj AllPers UsageRate ( j , tj , r , tr ) · PordVol ( j , tj )
  • TotShipVol (r, tr) is the total amount of resource r that is shipped in period tr (“total shipment volume”). TotProdVol (r, tr) is the total amount of resource r that is produced in period tr (“total production volume”). ConsVol (r, tr) is the amount of resource r that is consumed in period tr (“consumption volume”). Note that, if r is not a product, then TotProdVol (r, tr)=0. A production plan is required to satisfy the following constraints:
  • ∀jεAllOpn, ∀tjεAllPer:
      • ProdVol (j, tj)≧0 [non-negativity constraints]
  • ∀dεAllDem, ∀tdεAllPer:
      • ShipVol (d, td)≧0 [non-negativity constraints]
  • ∀rεAllMat, ∀trεAllPer:
      • StockVol (r, tr)≧0 [non-negativity constraints]
  • ∀rεAllCap, ∀trεAllPer:
      • StockVol (r, tr)=0 [non-stocking constraints]
  • ∀rεAllRes, ∀trεAllPer:
      • ConsSupVol (r, tr)≧0 [non-negativity constraints]
  • ∀rεAllRes, ∀trεAllPer:
      • ConsSupVol (r, tr) [SupplyVol (r, tr) [supply constraints]
  • ∀dεAllDem, ∀tdεAllPer: td = 1 td ShipVol ( d , td ) td = 1 td DemandVol ( d , td )
  • [demand constraints]
  • ∀rεAllRes, ∀trεAllPer:
      • TotShipVol (r, tr)+ConsVol (r, tr)+StockVol (r, tr)
      •  ≦ConsSupVol (r, tr)+TotProdVol (r, tr)+
      •  StockVol (r, tr-1) [resource constraints]
        A production plan that satisfies all of the above constraints is said to be feasible.
  • An important special case of the RCPP problem is the Positive Consumption RCPP problem. An RCPP problem is defined to be a Positive Consumption RCPP (PC RCPP) problem, if it has the following property:
  • ∀jεAllOpn, tjεAllPer, ∀rεBOR (j), trεAllPer:
  • UsageRate (j, tj, r, tr)≧0
  • Note that this implies the following property:
  • ∀jεAllOpn, ∀rεBOR (j):
  • ResIdx (ProdRes (j))>ResIdx (r).
  • Precise Definition of Pegging
  • Given an RCPP problem and feasible production plan for it, a pegging of the production plan has three portions: (1) a production pegging, (2) a stock pegging, and a (3) supply pegging. Given any operation j, any period tj, any demand d, and any period td, a production pegging specifies an associated quantity, PgdProdVol (j, tj, d, td) (“pegged production volume”). This is the portion of ProdVol (j, tj) that is pegged to the shipment of demand d in period td.
  • A production pegging must satisfy the following two constraints:
  • ∀jεAllOpn, ∀tjεAllPer, ∀dεAllDem, ∀tdεAllPer:
      • PgdProdVol (j, tj, d, td)≧0 [non-negativity constraints]
  • ∀jεAllOpn, ∀tjεAllPer: d AllDems td AllPers PgdProdVol ( j , tj , d , td ) ProdVol ( j , tj ) [ total pegging constraints ]
    Similarly, given any resource r, any period tr, any demand d, and any period td, a stock pegging specifies an associated quantity, PgdStockVol (r, tr, d, td) (“pegged stock volume”). This is the portion of StockVol (r, tr) that is pegged to the shipment of demand d in period td.
  • A stock pegging must satisfy the following two constraints:
  • ∀rεAllRes, ∀trεAllPer, ∀dεAllDem, ∀tdεAllPer:
      • PgdStockVol (r, tr, d, td)≧0 [non-negativity constraints]
  • ∀rεAllRes, ∀trεAllPer: d AllDems td AllPers PgdStockVol ( r , tr , d , td ) StockVol ( r , tr ) [ total pegging constraints ]
    Note that this implies that PgdStockVol (r, tr, d, td)=0, if r is a capacity.
  • Finally, given any resource r, any period tr, any demand d, and any period td, a supply pegging specifies an associated quantity, PgdConsSupVol (r, tr, d, td) (“pegged consumed supply volume”). This is the portion of ConsSupVol (r, tr) that is pegged to the shipment of demand d in period td.
  • A supply pegging must satisfy the following two constraints:
  • ∀rεAllRes, ∀trεAllPer, ∀dεAllDem, ∀tdεAllPer:
      • PgdConsSupVol (r, tr, d, td)≧0 [non-negativity constraints]
  • ∀rεAllRes, ∀trεAllPer: d AllDems td AllPers PgdConsSupVol ( r , tr , d , td ) ConsSupVol ( r , tr )
  • [total pegging constraints]
  • A pegging of a production plan for an RCPP problem is simply a production pegging, stock pegging, and supply pegging that satisfy the above six constraints. The total pegging constraints enforce a requirement that any quantity of production, stock, or supply that is pegged to one shipment cannot be pegged to any other. Note that the total pegging constraints are inequality constraints. This reflects the fact that it is acceptable for some portion of the production, stock, and supply not to be pegged to any shipment.
  • Plausible Peggings: The “Feasible Partitioning” Property
  • Given an RCPP problem, a feasible production plan for it, and a pegging of the production plan, and given any demand, d* and any period, td*, the pegged production plan for (d*, td*) is defined as follows:
      • ShipVolP (d*, td*)=ShipVol (d*, td*)
      • ∀dεAllDem, ∀tdεAllPer:
        • If d≠d* or td≠td*, then ShipVolP (d, td)=0
      • ∀jεAllOpn, ∀tjεAllPer:
        • ProdVolP (i, tj)=PgdProdVol (j, tj, d*, td*)
      • ∀rεAllRes, ∀trεAllPer:
        • StockVolp (r, tr)=PgdStockVol (r, tr, d*, td*)
        • ConsSupVolp (r, tr)=PgdConsSupVol (r, tr, d*, td*)
          The pegging is defined to be a feasible partitioning of the production plan if, for each demand d* and each period td*, the pegged production plan for (d*, td*) is feasible. Informally, this means that the pegging has partitioned the production plan into a set of disjoint portions associated with each shipment, where each portion defines a feasible production plan for achieving its corresponding shipment. This is a precise way of saying that the pegging is plausible.
  • In the context of a Positive Consumption RCPP problem, the pegging constructed by the PBE technique is guaranteed to be a feasible partitioning.
  • The Shipment Sequence
  • In addition to an RCPP problem and a production plan, the PBE technique optionally accepts one extra item of input data: a shipment sequence, ShipSeq. This is an ordered list of triples (d, td, IncShipVol),where dεAllDem, tdεAllPer, and IncShipVol≧0 (“incremental shipment volume”). The shipment sequence is required to satisfy the following constraints: d AllDem , td AllPer : ( d , td , IncShipVol ) ShipSeq IncShipVol = ShipVol ( d , td )
    In essence, a shipment sequence is simply the set of all shipments in the production plan arranged into some specific order, where the shipments may be broken into partial shipments that can be interlaced within the sequence.
  • The shipment sequence is optional input, because the PBE technique can easily generate one by default. For a default shipment sequence, it can simply go through the periods in order, and for each period, it can go through the demands in any order:
    {(demand #1, period #1, shipVol (demand #1, period #1)),
     (demand #2, period #1, shipVol (demand #2, period #1)),
     ...,
     (demand #1, period #2, shipVol (demand #1, period #2)),
    ...}
  • By specifying a non-default shipment sequence, the user can exercise some control over the pegging that will result: The resources that the PBE technique pegs first will be pegged to the shipments that appear earliest in the shipment sequence.
  • The data flow for the PBE technique with an optional shipment sequence is illustrated in FIG. 2. Here, blocks 6 to 10 correspond to blocks 1 to 5 of FIG. 1, while block 11 shows how the optional shipment sequence relates to the overall data flow.
  • The Pegging By Explosion Technique for the Positive Consumption RCPP Problem
  • The Pegging By Explosion technique will initially be described in the context of the PC RCPP problem. Following this initial description, the technique will be extended to apply to the fully general RCPP problem.
  • The PBE technique functions by reconstructing the production plan in a somewhat modified form. Let ProdVolR ( ), StockVolR ( ), ConsSupVolR ( ), and ShipVolR ( ) denote the reconstructed production plan. Upon conclusion of the reconstruction, the following properties hold:
      • ∀dεAllDem, ∀tdεAllPer:
        ShipVolR (d, td)=ShipVol (d, td)  (1)
      • ∀jεAllOpn, ∀tjεAllPer:
        ProdVolR (j, tj)≦ProdVol (j, tj)   (2)
      • ∀rεAllRes, ∀trεAllPer:
        StockVolR (r, tr)≦StockVol (r, tr)   (3)
        ConsSUPVolR (r, tr)≦ConsSupVol (r, tr)   (4)
        In addition, the reconstructed production plan is feasible. In other words, the reconstructed plan achieves the same set of shipments as the original plan, by only using production, stock, and supply that was used in the original.
  • The description of the PBE algorithm will be broken down into smaller “procedures”. The most important procedure in the PBE algorithm is called “Explode-For-PBE”. This procedure takes as its argument a triple (d*, td*, IncShipVol) from the shipment sequence and computes an incremental production plan for the triple, denoted by ProdVolI ( ), StockVolI ( ), ConsSupVolI ( ), and ShipVolI ( ).
  • When the Explode-For-PBE procedure returns, the following properties hold:
      • ∀dεAllDem, ∀tdεAllPer:
        If d=d* and td=td*, then ShipVolI (d, td)=IncShipVol   (5)
        Otherwise, ShipVolI (d, td)=0   (5)
      • ∀jεAllOpn, ∀tjεAllPer:
        ProdVolR (j, tj)+ProdVolI (j, tj)≦ProdVol (j, tj)  (6)
      • ∀rεAllRes, ∀trεAllPer:
        StockVolR (r, tr)+StockVolI (r, tr)≦StockVol (r, tr)   (7)
        ConsSupVolR (r, tr)+ConsSupVolI (r, tr)−ConsSupVol (r, tr)   (8)
        The incremental production plan is feasible.   (9)
        Note: In the following description, the symbol “←” represents assignment. Thus, “x←y” means “x is set to the value of y”. Also, the symbol “←+” represents adding to, so that “x←+y” means “x←x+y”.
  • The top-level logic of the PBE technique can be described as follows:
  • Algorithm Pegging-By-Explosion:
  • [Initialize the reconstructed data and pegging data to zero]
  • ∀dεAllDem, ∀tdεAllPer, ∀jεAllOpn, ∀tjεAllPer, ∀rεAllRes, ∀trεAllPer:
    ShipVolR (d, td)
    Figure US20060069597A1-20060330-P00801
    0
    ProdVolR (j, tj)
    Figure US20060069597A1-20060330-P00801
    0
    StockVolR (r, tr)
    Figure US20060069597A1-20060330-P00801
    0
    ConsSupVolR (r, tr)
    Figure US20060069597A1-20060330-P00801
    0
    PgdProdVol (j, tj, d, td)
    Figure US20060069597A1-20060330-P00801
    0
    PgdStockVol (r, tr, d, td)
    Figure US20060069597A1-20060330-P00801
    0
    PgdConsSupVol (r, tr, d, td)
    Figure US20060069597A1-20060330-P00801
    0

    For each triple (d*, td*, IncShipVol)εShipSeq, taken in sequence order, do the following:
  • Call Explode-For-PBE (d*, td*, IncShipVol).
      • [Add the incremental production plan to the reconstructed production plan.]
  • ∀dεAllDem, ∀tdεAllPer:
      • ShipVolR (d, td)←+ShipVolI (d, td)
  • ∀jεAllOpn, ∀tjεAllPer:
      • ProdVolR (j, tj)←+ProdVolI (j, tj)
  • ∀rεAllRes, ∀trεAllPer:
      • StockVolR (r, tr)←+StockVolI (r, tr)
      • ConsSupVolR (r, tr)←+ConsSupVolI (r, tr)
      • [Peg the incremental production, stock, and consumed supply volumes to the current demand.]
  • ∀jεAllOpn, ∀tjεAllPer:
      • PgdProdVol (, tj, d*, td*)←+ProdVolI (j, tj)
  • ∀rεAllRes, ∀trεAllPer:
      • PgdStockVol (r, tr, d*, td*)←+StockVolI (r, tr)
      • PgdConsSupVol (r, tr, d*, td*)←+ConsSupVolI (r, tr)
        Report the pegging: PgdProdVol ( ), PgdStockVol ( ), PgdConsSupVol ( ). Stop.
        Assuming that the Explode-For-PBE procedure achieves conditions (5), (6), (7), (8), and (9), it is easy to see that, upon completion of the PBE technique, the reconstructed production plan satisfies conditions (1), (2), (3), and (4) above and is feasible. It is also easy to see that PgdProdVol ( ), PgdStockVol ( ), and PgdConsSupVol ( ) form a valid pegging of the original production plan and that this pegging is a feasible partitioning of the production plan. It remains to describe the Explode-For-PBE procedure.
  • FIG. 3 illustrates the top level logic of the PBE algorithm implemented in computer software. Upon starting the procedure at start block 12, a PC RCPP problem, a feasible production plan, and a shipment sequence, ShipSeq, are read in at function block 13. In initialization block 14, the reconstructed production is set to zero and pegging is set to zero. Then, processing block 15 for ∀(d*, td*, IncShipVol)εShipSeq in order is entered. This processing block iterates through the shipment sequence. The Explode-Triple (d*, td*, IncShipVol) process is called in function block 16. In function block 17, the incremental production plan is added to the reconstructed production plan. Finally, the incremental production plan is pegged to the shipment (d*, td*) in function block 18 to complete the processing in processing block 15. Once this processing is complete, the pegging is reported in function block 19, and the process stops in stop block 20.
  • FIG. 4 illustrates the top level logic of the Explode-Triple procedure of block 16 in FIG. 3. Upon this process being called at start block 21, initialization block 22 sets the incremental production plan to zero and sets the ReqVol ( ) to zero. Then, in function block 23, ShipVolI (d*, td*) is set to IncShipVol and ReqVol (DemRes (d*, td*) is set to IncShipVol. At this point, nested processing block 24 ∀rεAllRes, in order of decreasing ResIndx(r), and processing block 25 ∀trεAllPer, in decreasing order, are entered. Processing block 25 calls the main-iteration (r, tr) processing block 26. Blocks 24 to 26 show the main iteration (block 26), as contained within the loop through all periods (block 25), as contained with a loop through all resources (block 24). When processing in block 24 and 25 are completed, a return is made to the main program shown in FIG. 3.
  • FIG. 5 illustrates the main iteration 26 of the Explode-Triple procedure of FIG. 4, treating it as a separate procedure for illustration purposes. When called at start block 28, a determination is made in decision block 29 as to whether ReqVol (r, tr)=0. If so, a return is made at block 30 to processing block 26; otherwise, net requirement is netted against supply in function block 31. Then, in decision block 32 a determination is made as to whether ReqVol (r, tr)=0. If so, a return is made at block 33 to processing block 26; otherwise, processing block 34 ∀jεProdOpns (r) is entered. The requirement to production is exploded by operation j in function block 35. A determination is made in decision block 36 as to whether ReqVol (r, tr)=0. If so, a return is made at block 37 to processing block 26; otherwise, the processing in processing block 34 continues to the next j in function block 38. Blocks 29, 30, 32, 33, 36, and 37 indicate the various opportunities for the iteration to terminate early. Blocks 34 to 38 indicate the loop through all operations that produce the resource. Blocks 31, 35 and 39 show how the three methods for fulfilling requirements fit into the main iteration. When the processing in block 34 completes, the requirement is exploded to stock in period t-1 in function block 39, and a return is made at block 40 to processing block 26.
  • The Explode-For-PBE procedure is similar to the explosion technique of Material Requirements Planning. The explosion functions by associating a quantity ReqVol (r, tr) (“requirement volume”) with each resource r in each period tr. The Explode-For-PBE procedure calls three more procedures, Net-Against-Supply, Explode-To-Production, and Explode-To-Stock. These four procedures can be described as follows:
  • Procedure Explode-For-PBE:
  • Given: A triple (d*, td*, IncShipVol)εShipSeq
  • ∀dεAllDem, ∀tdεAllPer:
      • ShipVolI (d, td)←0
  • ∀jεAllOpn, ∀tjεAllPer:
      • ProdVolI (j, tj)←0
  • ∀rεAllRes , ∀trεAllPer:
      • StockVolI (r, tr)←0
      • ConsSupVolI (r, tr)←0
      • ReqVol (r, tr)←0
  • ShipVolI (d*, td*)←IncShipVol
  • ReqVol (DemRes (d*), td*)←IncShipVol
  • Do the following for each resource rεAllRes in order of decreasing ResIdx (r):
  • Do the following main iteration for each period trεAllPer in decreasing order:
      • If ReqVol (r, tr)=0, skip this main iteration.
      • Call Net-Against-Supply (r, tr).
      • If ReqVol (r, tr)=0, terminate this main iteration.
      • Do the following for each operation jεProdOpns (r):
        • Call Explode-To-Production (r, j, tr).
        • If ReqVol (r, tr)=0, terminate this main iteration.
      • Call Explode-To-Stock (r, tr).
        Return.
        Procedure Net-Against-Supply:
        [Net the requirement against consumed supply.]
        Given: rεAllRes, trεAllPer.
        ConsSupVolI (r, tr)←Min ((ConsSupVol (r, tr)−ConsSupVolR (r, tr)), ReqVol (r, tr)) ReqVol (r, tr)←+−ConsSupVolI (r, tr)
        Return.
        Procedure Explode-To-Production:
        [Explode the requirement to production by operation j.]
        Given: rεAllRes, jεProdOpns (r), trεAllPer.
        ProdVolI (j, tr)←Min ((ProdVol (j, tr)−ProdVolR (r, tr)), ReqVol (r, tr))
        ∀r′εBOR (j):
  • ∀tr′εAllPer:
      • ReqVol (r′, tr′)+UsageRate (j, tr, r′, tr′)*ProdVolI (j, tr)
  • ReqVol (r, tr)←+−ProdVolI (j, tr)
  • Return.
  • Procedure Explode-To-Stock:
  • [Explode the requirement to stock from period tr-1.]
  • Given: rεAllRes, trεAllPer.
  • If tr=1, return.
  • StockVolI (r, tr-1) Min ((StockVol (r, tr-1)−StockVolR (r, tr-1)), ReqVol (r, tr))
  • ReqVol (r, tr-1)←+StockVolI (r, tr-1)
  • ReqVol (r, tr)←+−StockVolI (r, tr-1)
  • Return.
  • It is easy to see that Explode-For-PBE computes an incremental production plan that satisfies conditions (5), (6), (7), and (8) above. It can also be shown that condition (9) is satisfied, by verifying that ReqVol(r, tr)=0 at the end of each main iteration. In MRP terms, this means that there are no “net requirements”. In fact, there are a number of major differences between the Explode-For-PBE procedure and a conventional MRP explosion:
      • Explode-For-PBE never has any net requirements.
      • Explode-For-PBE can explode a requirement to multiple alternative operations for producing the same product.
      • Explode-For-PBE can explode a requirement to production in an earlier period (by exploding to stock).
        All of these differences are enabled by exploiting knowledge of the pre-existing production plan.
  • The main iteration of Explode-For-PBE fills the requirement, ReqVol (r, tr), using three alternative procedures: Net-Against-Supply, Explode-To-Production, and Explode-To-Stock. The order in which these three procedures are called is arbitrary: they can actually be called in any order. Also, the calls to Explode-To-Production for each operation in ProdOpns (r) can be done in any order, and indeed, these calls can be interlaced with the calls to Net-Against-Supply and Explode-To-Stock in any order. In general, using different alternative orderings for these procedure calls will result in different but equally valid peggings. Using any such ordering is considered to be a version of the PBE technique.
  • Depending on the application, there may be business reasons for choosing one ordering instead of another. For example, the ordering used in the algorithm description above has the effect of pegging production in the latest possible period to the current shipment. Alternatively, calling Explode-To-Stock before Explode-To-Production would have the effect of pegging production in the earliest possible period to the current shipment.
  • Extension to the General RCPP Problem
  • The general RCPP problem is allowed to have negative consumption. Unfortunately, when negative is consumption is allowed, it is easy to construct examples of RCPP problems and production plans such that no feasible partitioning of the production plan exists. Thus the PBE technique for PC RCPP problems, which is designed to construct a feasible partitioning, cannot be expected to function properly in the presence of negative consumption. Given an RCPP problem with negative consumption and a corresponding feasible production plan, consider the following definitions:
      • ∀jεAllOpn, ∀tjεAllPer, ∀rεBOR (j), ∀trεAllPer:
        • If UsageRate (j, tj, r, tr)<0:
          • Let NegUsageRate (j, tj, r, tr)=−UsageRate (j, tj, r, tr)
        • If UsageRate (j, tj, r, tr)≧0:
          • Let NegUsageRate (j, tj, r, tr)=0
      • ∀rεAllRes, ∀trεAllPer: Let NegConsVol _ ( r , tr ) = j ConsOpns ( r ) tj AllPers NegUsageRate ( j , tj , r , tr ) * ProdVol ( j , tj )
        Thus, NegConsVol (r, tr) is the total negative consumption of resource r in period tr.
  • To see how the PBE technique is extended to handle the negative consumption case, define the auxiliary problem and auxiliary production plan, by making the following modifications to the original RCPP problem and production plan:
  • ∀rεAllRes:
  • Add an “auxiliary” operation, j*.
  • Let ProdRes (j*)=r.
  • Let BOR (j*)=Ø.
  • ∀trεAllPer:
      • Let ProdVol (j*, tr)=NegConsVol (r, tr).
        ∀jεAllOpn, ∀tjεAllPer, ∀rεBOR (j), ∀trεAllPer:
  • If UsageRate (j, tj, r, tr)<0:
      • UsageRate (j, tj, r, tr)←0
        In other words, the negative consumption is replaced by production from auxiliary operations and the negative usage rates are set to zero. The auxiliary problem and production plan have two important properties:
      • The auxiliary problem is a Positive Consumption RCPP problem.
      • The auxiliary production plan is feasible for the auxiliary problem.
        The PBE technique can be extended to handle negative consumption in either of the following two ways:
      • Explicit transformation: Explicitly construct the auxiliary problem and production plan, apply the PBE technique to them, and then transfer the resulting pegging back into the original problem and production plan.
      • Implicit transformation: Work with the original problem and production plan, by adding steps to the PBE technique that are mathematically equivalent to what the PBE technique would be doing with the auxiliary problem and production plan.
        More specifically, the implicit transformation approach makes the following modifications to the original algorithm. In the Explode-To-Production procedure, any value of UsageRate(j, tr, r′, tr′) that is negative is treated as zero. In addition to pegging the production volumes, etc., it also explicitly pegs the negative consumption volumes, by working with PgdNegConsVol (r, tr, d, td), NegConsVolR (r, tr), and NegConsVolI (r, tr). A call to the following procedure is added (in any order) to the Explode-For-PBE procedure:
        Procedure Explode-To-Negative-Consumption:
        Given: rεAllRes, trεAllPer.
        NegConsVolI (r, tr)←Min ((NegConsVol (r, tr)−NegConsVolR (r, tr)), ReqVol (r, tr))
        ReqVol (r, tr)←+−NegConsVolI (r, tr)
        Return.
  • The pegging constructed by the PBE technique in the negative consumption case is a feasible partitioning of the auxiliary production plan with respect to the auxiliary problem. This means that the pegging has partitioned the original production plan into a set of disjoint portions associated with each shipment, where each portion defines a feasible production plan for achieving its corresponding shipment, if negative consumption is treated as production by an auxiliary operation. This is a precise way of saying that the pegging is plausible in the presence of negative consumption.
  • Substitution
  • The definition of the RCPP problem can be extended to include substitution in the BOR structure. Specifically, given any operation j, for any resource rεBOR (j), an RCPP problem with substitution may specify a set of substitute resources, any of which can be consumed in place of r, and with potentially different usage rates, etc. For a detailed formulation of the RCPP problem with substitution, see U.S. Pat. No. 5,630,070.
  • Any RCPP problem with substitution can be transformed into an equivalent RCPP problem without substitution by the following technique. For each operation j and each resource rεBOR (j) that has substitution, modify to the problem as follows:
  • Add an extra capacity resource, r*.
  • Add r* to BOR (j).
  • Add an extra operation, j*.
  • Let ProdRes (j*)=r*.
  • Let BOR(j*)=r.
  • For each substitute resource r′ associated with j and r, do the following:
  • Add an extra operation, j**.
  • Let ProdRes (j**)=r*.
  • Let BOR (j**)=r′.
  • Remove r from BOR (j).
  • Also, the usage rate data for the new objects is copied from the usage rates in the original data as appropriate. This transformed problem is equivalent to the original problem in the sense that it leads to an equivalent set of feasible production plans. Also, the transformed problem has no substitution. The PBE technique can be extended to handle substitution either by using an explicit transformation approach or by using an implicit transformation approach, with respect to the transformation defined above.
  • Pegging Infeasibilities
  • Some RCPP systems allow the production plan that they produce to be infeasible. In particular, the resource constraints are sometimes allowed be to violated, perhaps with a penalty in an objective function. The PBE technique can be extended to cover this case, by using a problem reduction approach. Given an RCPP problem and a production plan that violates some of the resource constraints, but is otherwise feasible, define the following transformed problem. For each resource r and each period tr whose resource constraint is violated, modify the problem and production plan as follows:
  • Let ViolVol (r, tr)=The amount by which the left-hand side of the resource constraint on tr and r exceeds the right-hand-side (“violation volume”).
  • Add an extra operation, j*.
  • Let ProdRes (j*)=r.
  • Let BOR (j*)=Ø.
  • Let ProdVol (j*, tr)=ViolVol (r, tr).
  • It is easy to see that the transformed production plan is feasible for the transformed problem.
  • This transformation can be used to extend the PBE technique to handle resource constraint violations by applying either an explicit transformation approach or an implicit transformation approach. In either case, the production pegging for the extra operations has special significance. For any demand d and any period td, PgdProdVol (j*, tr, d, td) is the amount of the violation of the resource constraint for resource r in period tr that is pegged to the shipment to demand d in period td. (It might be notated as PgdViolVol (r, tr, d, td).)
  • Note that, when using the implicit approach, the use of ViolVol (r, tr) to fill the requirement ReqVol (r, tr) in the Explode-For-PBE procedure can be done in any order compared to the other procedures for filling the requirement (Net-To-Supply, Explode-To-Production, Explode-To-Stock, and Explode-To-Negative-Consumption). Normally, it would be done last, since violating a resource constraint would normally be done only as a last resort. A special case of RCPP where resource constraints can be violated is MRP. Thus the above techniques can be used to do pegging of a pre-existing MRP solution.
  • While the invention has been described in terms of preferred embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.

Claims (30)

1. A computer implemented method for constructing a pegging of a production plan for a resource-constrained production planning (RCPP) problem, comprising the step of constructing the pegging after the production plan has already been generated.
2. The computer implemented method recited in claim 1, wherein the step of constructing the pegging comprises the steps of:
reconstructing a modified version of the production plan by an incremental allocation technique, and
then applying a Pegging By Monitoring technique to the reconstructed production plan in order to produce the pegging.
3. The computer implemented method recited in claim 2, wherein the step of reconstructing a modified verison of the production plan is accomplished by a series of “explosion” steps similar to an explosion technique of Material Requirements Planning (MRP).
4. The computer implemented method recited in claim 3, wherein the method is restricted to be used only on Positive Consumption resource-constrained production planning (RCPP) problems and comprises of the steps of:
reading in a Positive Consumption RCPP problem,
reading in a feasible production plan for the problem,
either reading in a shipment sequence corresponding to the shipments in the production plan, or computing a default shipment sequence,
initializing the reconstructed production plan to zero,
initializing the pegging to zero,
proceeding through the triples (d*, td*, IncShipVol) in the shipment sequence, with each demand d* in period td*, and performing the following three steps:
employing a specialized explosion technique to construct an incremental production plan (denoted ProdVolI ( ), StockVolI ( ), ConsSupVolI ( ), and ShipVolI ( )) that specifies a feasible means of enabling a shipment of IncShipVol units to demand d* in period td* without using any resources that have already been pegged,
adding the incremental production plan to the reconstructed production plan, and
pegging the incremental production plan to the current shipment (d*, td*), and
reporting the pegging.
5. The computer implemented method recited in claim 4, wherein the specialized explosion technique comprises the steps of:
initializing the incremental production plan to zero,
associating a quantity, ReqVol (r, tr), with each resource r in each period tr,
initializing all values of ReqVol (r, tr) to zero,
setting ShipVolI (d*, td*) to IncShipVol,
setting ReqVol (DemRes (d*), td*) to IncShipVol,
proceeding through each resource rεAllRes in order of decreasing ResIdx (r) and performing the following five steps:
proceeding through each period trεAllPer in decreasing order and performing the following four steps:
netting the requirement against supply,
∀jεProdOpns (r), performing the following step:
exploding the requirement to production by operation j, and
exploding the requirement to stock in period tr-1.
6. The computer implemented method recited in claim 5, wherein the step netting the requirement against supply comprises the steps of:
setting ConsSupVolI (r, tr)←Min ((ConsSupVol (r, tr)−ConsSupVolR (r, tr)), ReqVol (r, tr)), and
setting ReqVol (r, tr)←+−ConsSupVolI (r, tr),
and wherein the step exploding the requirement to production by operation j comprises the steps of:
setting ProdVolI (j, tr)←Min ((ProdVol (j, tr)−ProdVolR (j, tr)), ReqVol (r, tr)),
∀r′εBOR (j), where BOR is Bill-of-Resources, performing the following three steps:
∀tr′εAllPer, performing the following step:
setting ReqVol (r′, tr′)+UsageRate (j, tr, r′, tr′)*ProdVolI (j, tr),
setting ReqVol (r, tr)←+−ProdVolI (j, tr), and
wherein the step exploding the requirement to stock in period tr-1 comprises the steps of:
if tr=1, doing nothing, but otherwise performing the following three steps:
setting StockVolI (r, tr-1)←Min ((StockVol (r, tr-1)−StockVolR (r, tr-1)), ReqVol (r, tr)),
setting ReqVol (r, tr-1)←+StockVolI (r, tr-1), and
setting ReqVol (r, tr)←+−StockVolI (r, tr-1).
7. The computer implemented method recited in claim 6, wherein the steps netting the requirement against supply, exploding the requirement to production by operation j, and exploding the requirement to stock in period tr-1 are performed in any order, where this method is called Pegging By Explosion Technique for Positive Consumption RCPP Problems, or PC-PBE technique.
8. The computer implemented method recited in claim 7 (the PC-PBE technique), used to construct a pegging of a production plan for any RCPP problem, by using the method within the steps of:
explicitly constructing an auxiliary problem and production plan identical to the original problem and production plan, except that
for each resource, an “auxiliary” operation is added that produces that resource and consumes no resources,
the negative consumption of each resource is replaced with production by the auxiliary operation for that resource, and
all negative usage rates are set to zero, and
applying the PC-PBE technique to the auxiliary problem and production plan, and
transferring the resulting pegging back into the original problem and production plan.
9. The computer implemented method recited in claim 7 (the PC-PBE technique). used to construct pegging of a production plan for any RCPP problem, by using the method within the the steps of:
conceptually defining the auxiliary problem and production plan identical to the original problem and production plan, except that
for each resource, an “auxiliary” operation is added that produces that resource and consumes no resources,
the negative consumption of each resource is replaced with production by the auxiliary operation for that resource, and
all negative usage rates are set to zero,
working with the original problem and production plan, and
performing a sequence of steps that are mathematically equivalent to those that the PC-PBE technique would be performing, if it were being applied to the auxiliary problem and production plan, whereby the method is called the Pegging By Explosion Technique for RCPP Problems, or the PBE technique.
10. The computer implemented method recited in claim 9 (the PBE technique). used to construct a pegging of a production plan for a variant of an RCPP problem that has been generalized so as to allow substitution in a Bill-of-Resources (BOR) structure, by using the method within the steps of:
conceptually defining a transformed RCPP problem and production plan that is equivalent to the original problem and production plan, but has no substitution,
working with the original problem and production plan, and
performing a sequence of steps that are mathematically equivalent to those that the PBE technique would be performing, if it were being applied to the transformed problem and production plan.
11. The computer implemented method recited in claim 9 (the PBE technique), used to construct a pegging of a production plan for a variant of an RCPP problem that has been generalized so as to allow, substitution in a Bill-of-Resources (BOR) structure and to permit violations of the resource constraints, by using the method within the steps of:
conceptually defining a transformed RCPP problem and production plan that is equivalent to the original problem and production plan, but has no substitution and does not violate any of the resource constraints,
working with the original problem and production plan, and
performing a sequence of steps that are mathematically equivalent to those that the PBE technique would be performing, if it were being applied to the transformed problem and production plan.
12. The computer implemented method recited in claim 11, specifically applied to construct a pegging of the solution of a Material Requirements Planning problem.
13. The computer implemented method recited in claim 7, wherein the method is restricted to be used only on problems in which Nperiods=1, where the method is made simpler to implement in the following ways:
by eliminating the concept of a period from all of the data and notation;
by eliminating the distinction between material resources and capacity resources;
by deleting all references to stock volume (StockVol ( )) and its pegging; and
by removing the exploding the requirement to stock step from the method.
14. The computer implemented method recited in claim 7, wherein the method is restricted to be used only on problems in which each resource can only be produced by at most one operation, where the method is made simpler to implement in the following ways:
by eliminating the concept of an operation from all of the data and notation, eliminating all data associated with an operation and replacing it by corresponding data associated with a product, so that, e.g., ProdVol (j, ij) for operation j is replaced by ProdVol (r, tr) for product resource r;
by eliminating the following two steps from method:
∀jεProdOpns (r), performing the following step:
exploding the requirement to production by operation j;
by replacing the above two steps with the following step:
exploding the requirement to production, if the resource is a product.
15. The computer implemented method recited in claim 7, wherein the method is restricted to be used only on problems in which Nperiods=1 and in which each resource can only be produced by at most one operation, where the method is made simpler to implement in the following ways:
by eliminating the concept of a period from all of the data and notation;
by eliminating the distinction between material resources and capacity resources;
by deleting all references to stock volume (StockVol ( )) and its pegging;
by removing the exploding the requirement to stock step from the method,
by eliminating the concept of an operation from all of the data and notation, eliminating all data associated with an operation and replacing it by corresponding data associated with a product, so that, e.g., ProdVol (j, tj) for operation j is replaced by ProdVol (r, tr) for product resource r;
by eliminating the following two steps from method:
∀jεProdOpns (r), performing the following step:
exploding the requirement to production by operation j;
by replacing the above two steps with the following step:
exploding the requirement to production, if the resource is a product.
16. A computer readable medium containing code implementing a method for constructing a pegging of a production plan for a resource-constrained production planning (RCPP) problem, said method comprising the step of constructing the pegging after the production plan has already been generated.
17. The computer readable medium recited in claim 16, wherein the step of constructing the pegging comprises the steps of:
reconstructing a modified version of the production plan by an incremental allocation technique, and
then applying a Pegging By Monitoring technique to the reconstructed production plan in order to produce the pegging.
18. The computer readable medium recited in claim 17, wherein the step of reconstructing a modified verison of the production plan is accomplished by a series of “explosion” steps similar to an explosion technique of Material Requirements Planning (MRP).
19. The computer readable medium recited in claim 18, wherein the method is restricted to be used only on Positive Consumption resource-constrained production planning (RCPP) problems and comprises of the steps of:
reading in a Positive Consumption RCPP problem,
reading in a feasible production plan for the problem,
either reading in a shipment sequence corresponding to the shipments in the production plan, or computing a default shipment sequence,
initializing the reconstructed production plan to zero,
initializing the pegging to zero,
proceeding through the triples (d*, td*, IncShipVol) in the shipment sequence, with each demand d* in period td*, and performing the following three steps:
employing a specialized explosion technique to construct an incremental production plan (denoted by ProdVolI ( ), StockVolI ( ), ConsSupVolI ( ), and ShipVolI ( ) that specifies a feasible means of enabling a shipment of IncShipVol units to demand d* in period td* without using any resources that have already been pegged,
adding the incremental production plan to the reconstructed production plan, and
pegging the incremental production plan to the current shipment (d*, td*), and
reporting the pegging.
20. The computer readable medium recited in claim 19, wherein the specialized explosion technique comprises the steps of:
initializing the incremental production plan to zero,
associating a quantity, ReqVol (r, tr), with each resource r in each period tr,
initializing all values of ReqVol (r, tr) to zero,
setting ShipVolI (d*, td*) to IncShipVol,
setting ReqVol (DemRes (d*), td*) to IncShipVol,
proceeding through each resource rεAllRes in order of decreasing ResIdx (r) and performing the following five steps:
proceeding through each period tr E AllPer in decreasing order and performing the following four steps:
netting the requirement against supply,
∀jεProdOpns (r), performing the following step:
exploding the requirement to production by operation j, and
exploding the requirement to stock in period tr-1.
21. The computer readable medium recited in claim 20, wherein the step netting the requirement against supply comprises the steps of:
setting ConsSupVolI (r, tr)←Min ((ConsSupVol (r, tr)−ConsSupVolR (r, tr)), ReqVol (r, tr)), and
setting ReqVol (r, tr)←+−ConsSupVolI (r, tr),
and wherein the step exploding the requirement to production by operation j comprises the steps of:
setting ProdVolI (I, tr)←Min ((ProdVol (I, tr)−ProdVolR (j, tr)), ReqVol (r, tr)),
∀r′εBOR (j), where BOR is Bill-of-Resources, performing the following three steps:
∀tr′εAllPer, performing the following step:
setting ReqVol (r′, tr′)+UsageRate (j, tr, r′, tr′)*ProdVolI (I, tr),
setting ReqVol (r, tr)←+−ProdVolI (j, tr), and
wherein the step exploding the requirement to stock in period tr-1 comprises the steps of:
if tr=1, doing nothing, but otherwise performing the following three steps:
setting StockVolI (r, tr-1)←Min ((StockVol (r, tr-1)−StockVolR (r, tr-1)), ReqVol (r, tr)),
setting ReqVol (r, tr-1)←+StockVolI (r, tr-1), and
setting ReqVol (r, tr)←+−StockVolI (r, tr-1).
22. The computer readable medium recited in claim 20, wherein the steps netting the requirement against supply, exploding the requirement to production by operation j, and exploding the requirement to stock in period tr-1 are performed in any order, where this method is called Pegging By Explosion Technique for Positive Consumption RCPP Problems, or PC-PBE technique.
23. The computer readable medium recited in claim 22 (the code implementing the PC-PBE technique) where in the method is used to construct a pegging of a production plan for any RCPP problem, by using the method within the steps of:
explicitly constructing an auxiliary problem and production plan identical to the original problem and production plan, except that
for each resource, an “auxiliary” operation is added that produces that resource and consumes no resources,
the negative consumption of each resource is replaced with production by the auxiliary operation for that resource, and
all negative usage rates are set to zero, and
applying the PC-PBE technique to the auxiliary problem and production plan, and
transferring the resulting pegging back into the original problem and production plan.
24. The computer readable medium recited in claim 22 (the code implementing the PC-PBE technique), wherein the method is used to construct a pegging of a production plan for any RCPP problem, by using the method within the steps of:
conceptually defining the auxiliary problem and production plan identical to the original problem and production plan, except that
for each resource, an “auxiliary” operation is added that produces that resource and consumes no resources,
the negative consumption of each resource is replaced with production by the auxiliary operation for that resource, and
all negative usage rates are set to zero,
working with the original problem and production plan, and
performing a sequence of steps that are mathematically equivalent to those that the PC-PBE technique would be performing, if it were being applied to the auxiliary problem and production plan, whereby the method is called Pegging By Explosion Technique for RCPP Problems, or the PBE technique.
25. The computer readable method medium recited in claim 24 (the code implementing the PBE technique), wherein the method is used to construct a pegging of a production plan for a variant of an RCPP problem that has been generalized so as to allow substitution in a Bill-of-Resources (BOR) structure, by using the method within the steps of:
conceptually defining a transformed RCPP problem and production plan that is equivalent to the original problem and production plan, but has no substitution,
working with the original problem and production plan, and
performing a sequence of steps that are mathematically equivalent to those that the PBE technique would be performing, if it were being applied to the transformed problem and production plan.
26. The computer readable medium recited in claim 24 (the code implementing the PBE technique), wherein the method is used to construct a pegging of a production plan for a variant of an RCPP problem that has been generalized so as to allow substitution in a Bill-of-Resources (BOR) structure and to permit violations of the resource constraints, by using the method within the steps of:
conceptually defining a transformed RCPP problem and production plan that is equivalent to the original problem and production plan, but has no substitution and does not violate any of the resource constraints,
working with the original problem and production plan, and
performing a sequence of steps that are mathematically equivalent to those that the PBE technique would be performing, if it were being applied to the transformed problem and production plan.
27. The computer readable medium recited in claim 26, specifically applied to construct a pegging of the solution of a Material Requirements Planning problem.
28. The computer readable medium recited in claim 22, wherein the method is restricted to be used only on problems in which Nperiods=1, where the method is made simpler to implement in the following ways:
by eliminating the concept of a period from all of the data and notation;
by eliminating the distinction between material resources and capacity resources;
by deleting all references to stock volume (StockVol 0) and its pegging; and
by removing the exploding the requirement to stock step from the method.
29. The computer readable medium recited in claim 22, wherein the method is restricted to be used only on problems in which each resource can only be produced by at most one operation, where the method is made simpler to implement in the following ways:
by eliminating the concept of an operation from all of the data and notation, eliminating all data associated with an operation and replacing it by corresponding data associated with a product, so that, e.g., ProdVol (j, tj) for operation j is replaced by ProdVol (r, tr) for product resource r;
by eliminating the following two steps from method:
∀jεProdOpns (r), performing the following step:
exploding the requirement to production by operation j;
by replacing the above two steps with the following step:
exploding the requirement to production, if the resource is a product.
30. The computer readable medium recited in claim 22, wherein the method is restricted to be used only on problems in which Nperiods=1 and in which each resource can only be produced by at most one operation, where the method is made simpler to implement in the following ways:
by eliminating the concept of a period from all of the data and notation;
by eliminating the distinction between material resources and capacity resources;
by deleting all references to stock volume (StockVol ( )) and its pegging;
by removing the exploding the requirement to stock step from the method.
by eliminating the concept of an operation from all of the data and notation, eliminating all data associated with an operation and replacing it by corresponding data associated with a product, so that, e.g., ProdVol (j, tj) for operation j is replaced by ProdVol (r, tr) for product resource r;
by eliminating the following two steps from method:
∀jεProdOpns (r), performing the following step:
exploding the requirement to production by operation j;
by replacing the above two steps with the following step:
exploding the requirement to production, if the resource is a product.
US10/953,181 2004-09-30 2004-09-30 Pegging resources to shipments in a pre-existing production plan Abandoned US20060069597A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US10/953,181 US20060069597A1 (en) 2004-09-30 2004-09-30 Pegging resources to shipments in a pre-existing production plan

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US10/953,181 US20060069597A1 (en) 2004-09-30 2004-09-30 Pegging resources to shipments in a pre-existing production plan
US12/061,801 US8326670B2 (en) 2004-09-30 2008-04-03 Pegging resources to shipments in a pre-existing production plan
US13/556,555 US20120290354A1 (en) 2004-09-30 2012-07-24 Pegging Resources to Shipments in a Pre-Existing Production Plan

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US12/061,801 Continuation US8326670B2 (en) 2004-09-30 2008-04-03 Pegging resources to shipments in a pre-existing production plan

Publications (1)

Publication Number Publication Date
US20060069597A1 true US20060069597A1 (en) 2006-03-30

Family

ID=36100380

Family Applications (3)

Application Number Title Priority Date Filing Date
US10/953,181 Abandoned US20060069597A1 (en) 2004-09-30 2004-09-30 Pegging resources to shipments in a pre-existing production plan
US12/061,801 Expired - Fee Related US8326670B2 (en) 2004-09-30 2008-04-03 Pegging resources to shipments in a pre-existing production plan
US13/556,555 Abandoned US20120290354A1 (en) 2004-09-30 2012-07-24 Pegging Resources to Shipments in a Pre-Existing Production Plan

Family Applications After (2)

Application Number Title Priority Date Filing Date
US12/061,801 Expired - Fee Related US8326670B2 (en) 2004-09-30 2008-04-03 Pegging resources to shipments in a pre-existing production plan
US13/556,555 Abandoned US20120290354A1 (en) 2004-09-30 2012-07-24 Pegging Resources to Shipments in a Pre-Existing Production Plan

Country Status (1)

Country Link
US (3) US20060069597A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070162070A1 (en) * 1997-11-07 2007-07-12 Salviac Limited Embolic protection device
US20090157474A1 (en) * 2007-12-18 2009-06-18 Sap Ag Managing Supply and Demand for a Ware
US20130085797A1 (en) * 2011-10-04 2013-04-04 Fabrice Laur Promise driven planning in a supply chain
US20140136282A1 (en) * 2012-02-13 2014-05-15 Joseph Fedele Method and Apparatus for Consumer Procurement Aggregation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5216593A (en) * 1991-01-24 1993-06-01 International Business Machines Corporation Method and apparatus for discrete activity resourse allocation through cardinality constraint generation
US5548518A (en) * 1994-05-31 1996-08-20 International Business Machines Corporation Allocation method for generating a production schedule
US5630070A (en) * 1993-08-16 1997-05-13 International Business Machines Corporation Optimization of manufacturing resource planning
US5971585A (en) * 1997-09-09 1999-10-26 International Business Machines Corporation Best can do matching of assets with demand in microelectronics manufacturing
US6272389B1 (en) * 1998-02-13 2001-08-07 International Business Machines Corporation Method and system for capacity allocation in an assembly environment

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0770967A3 (en) * 1995-10-26 1998-12-30 Philips Electronics N.V. Decision support system for the management of an agile supply chain
US6049742A (en) * 1997-09-26 2000-04-11 International Business Machines Corporation Projected supply planning matching assets with demand in microelectronics manufacturing
US7216086B1 (en) * 2001-04-30 2007-05-08 Cisco Technology, Inc. Method and apparatus providing a supply chain management system useful in outsourced manufacturing
JP2003091309A (en) * 2001-07-11 2003-03-28 Class Technology Co Ltd Production management system and production management method
US6701201B2 (en) * 2001-08-22 2004-03-02 International Business Machines Corporation Decomposition system and method for solving a large-scale semiconductor production planning problem
JP3618720B2 (en) * 2002-02-20 2005-02-09 イーマニファクチャリング株式会社 Management method based on flow management in the supply chain product management
US7529686B2 (en) * 2002-10-16 2009-05-05 International Business Machines Corporation Supply planning system and method utilizing constrained and unconstrained explosion and implosion of flagged demand statements
US20050137919A1 (en) * 2003-12-19 2005-06-23 International Business Machine Corporation Method, system, and storage medium for integrating return products into a forward supply chain manufacturing process
US7590461B2 (en) * 2006-04-06 2009-09-15 International Business Machines Corporation Large scale supply planning

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5216593A (en) * 1991-01-24 1993-06-01 International Business Machines Corporation Method and apparatus for discrete activity resourse allocation through cardinality constraint generation
US5630070A (en) * 1993-08-16 1997-05-13 International Business Machines Corporation Optimization of manufacturing resource planning
US5548518A (en) * 1994-05-31 1996-08-20 International Business Machines Corporation Allocation method for generating a production schedule
US5971585A (en) * 1997-09-09 1999-10-26 International Business Machines Corporation Best can do matching of assets with demand in microelectronics manufacturing
US6272389B1 (en) * 1998-02-13 2001-08-07 International Business Machines Corporation Method and system for capacity allocation in an assembly environment

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070162070A1 (en) * 1997-11-07 2007-07-12 Salviac Limited Embolic protection device
US20090157474A1 (en) * 2007-12-18 2009-06-18 Sap Ag Managing Supply and Demand for a Ware
US8355939B2 (en) * 2007-12-18 2013-01-15 Sap Ag Managing supply and demand for a ware
US20130085797A1 (en) * 2011-10-04 2013-04-04 Fabrice Laur Promise driven planning in a supply chain
US20140136282A1 (en) * 2012-02-13 2014-05-15 Joseph Fedele Method and Apparatus for Consumer Procurement Aggregation

Also Published As

Publication number Publication date
US20120290354A1 (en) 2012-11-15
US20080255913A1 (en) 2008-10-16
US8326670B2 (en) 2012-12-04

Similar Documents

Publication Publication Date Title
Yano Setting planned leadtimes in serial production systems with tardiness costs
Smith et al. Management of multi-item retail inventory systems with demand substitution
Khouja et al. A review of the joint replenishment problem literature: 1989–2005
McDaniel et al. A modified Benders' partitioning algorithm for mixed integer programming
US6963847B1 (en) System and method for managing ATP data in a distributed supply chain planning environment
US6233493B1 (en) Computer-implemented product development planning method
CA2195813C (en) System and method for extended enterprise planning across a supply chain
Wolsey Solving multi-item lot-sizing problems with an MIP solver using classification and reformulation
Talbot Resource-constrained project scheduling with time-resource tradeoffs: The nonpreemptive case
Vanderbeck Exact algorithm for minimising the number of setups in the one-dimensional cutting stock problem
Chundi et al. Deferred updates and data placement in distributed databases
US8027860B2 (en) Systems and methods for planning demand for configurable products
Billington et al. Mathematical programming approaches to capacity-constrained MRP systems: Review, formulation and problem reduction
US7318039B2 (en) Project risk management system utilizing probability distributions
US7835952B2 (en) System, method, and computer program product for creating a production plan
US7006981B2 (en) Assortment decisions
US7080026B2 (en) Supply chain demand forecasting and planning
Feng et al. Perishable asset revenue management with Markovian time dependent demand intensities
Önüt et al. A two-phase possibilistic linear programming methodology for multi-objective supplier evaluation and order allocation problems
US20020072958A1 (en) Residual value forecasting system and method thereof, insurance premium calculation system and method thereof, and computer program product
US20020095307A1 (en) System and method for inventory and capacity availability management
Chen et al. Analysis of relaxations for the multi-item capacitated lot-sizing problem
Stecke et al. Production and transportation integration for a make-to-order manufacturing company with a commit-to-delivery business mode
Muckstadt et al. Analysis of multistage production systems
US6910017B1 (en) Inventory and price decision support

Legal Events

Date Code Title Description
AS Assignment

Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:WITTROCK, ROBERT JAMES;REEL/FRAME:015445/0074

Effective date: 20040929