CN118822212A - Supply chain intelligent planning method, system, electronic device and storage medium - Google Patents

Abstract

The present invention discloses a supply chain intelligent planning method, system, electronic device and storage medium, and relates to the technical field of supply chain chain scheduling. The method includes: determining demand sorting according to orders; using product structure model and allocation constraints, formulating demand plans one by one according to the demand sorting, and obtaining supply chain product production planning scheme; the product structure model is composed of MLBOM multi-layer product structure; the allocation constraint is based on the constraint quantity based on time unit, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraint includes the output that the production line can provide per day, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate per day. The present invention can integrate internal and external resources and maximize the use of intelligent guidance for replenishment, transportation and production.

Description

Supply chain intelligent planning method, system, electronic device and storage medium

Technical Field

The present invention relates to the technical field of supply chain scheduling, and in particular to a supply chain intelligent planning method, system, electronic equipment and storage medium.

Background Art

In the current intelligent era, suppliers and brand owners need to give delivery dates for orders quickly and accurately, but the current challenges are: existing solutions can no longer meet the needs of supply chain management. Although intelligent integrated planning solutions have become a research and development direction, there is no corresponding technical support. With the development of industry, the cloud model of European and American systems and their inherent model algorithms can no longer support the needs of Chinese companies.

Summary of the invention

The purpose of the present invention is to provide a supply chain intelligent planning method, system, electronic device and storage medium, which can integrate internal and external resources and maximize the use of intelligent guidance for replenishment, transportation and production.

To achieve the above object, the present invention provides the following solutions:

A supply chain intelligent planning method, comprising:

Determine the order of demand based on the order;

By using the product structure model and allocation constraints, demand plans are formulated one by one according to the demand ranking to obtain the supply chain product production planning scheme; the product structure model is composed of an MLBOM multi-layer product structure; the allocation constraints are based on the constraint quantity based on the time unit, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraints include the output that the production line can provide per day, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate per day.

Optionally, the process of formulating the demand plan is:

In LTR Forward, the availability date of each production node is determined, and a delivery plan is formulated based on each availability date. Based on the delivery date, RTL Backward uses the constraint model to allocate resources layer by layer to generate planned orders. LTR Forward, or Loof To Root Forward, looks at the supply and demand matching from the leaf node to the root node from the current date to the future; RTL Backward, or Root To Leaf Backward, looks at the supply and demand matching from the root node to the leaf node from the future to the present.

Optionally, the MLBOM multi-layer product structure is a supply chain network transformed based on a single-layer BOM structure; in the supply chain network, the starting point is the ROOT root, COM is the child node, BOM is the parent node of the node, the node as the starting point has no parent node, ROOTLT is the total lead time from a certain node to ROOT, ROOTPER is how many units of nodes can be converted into one unit of ROOT, and the node without child nodes is Leaf.

Optionally, the single-layer BOM structure includes a parent node BOM and a child node COM; in the supply process, when flowing from any node to another node, the any node is the child node COM, and the other node is the parent node BOM.

Optionally, in the MLBOM multi-layer product structure, when a material source corresponds to multiple constraints, if they are in the same group, they are substitutable relationships; if they are not in the same group or the group is empty, they are combined relationships, which means that multiple combined relationships need to complete the same amount of materials in the same period, and the minimum value is taken using Rate/Factor. Rate refers to the capacity of a time unit (day, week, month, etc.), and Factor refers to the constraint consumed per unit of supply; Rate/Factor refers to the time unit consumed by the supply.

The present invention also provides a supply chain intelligent planning system, comprising:

A sorting determination unit, used to determine the demand sorting according to the order;

The supply planning unit is used to use the constraint model to formulate demand plans one by one according to the demand ranking to obtain the supply chain product production planning scheme; the constraint model is based on the constraint quantity based on the time unit, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraint includes the output that the production line can provide every day, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate every day; each Supply represents the flow process of goods, services or resources provided by a certain link in the supply chain to the next link; each unit Supply represents the specific quantity of goods, services or resources provided by a certain link in the supply chain to the next link.

The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above-mentioned supply chain intelligent planning method.

The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the supply chain intelligent planning method as described above.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

The present invention discloses a supply chain intelligent planning method, system, electronic device and storage medium, the method includes determining demand sorting according to orders; using product structure model and allocation constraints, formulating demand plans one by one according to the demand sorting, and obtaining supply chain product production planning scheme; the product structure model is composed of MLBOM multi-layer product structure; the allocation constraints are based on the number of constraints based on time units, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraint includes the output that the production line can provide per day, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate per day. The present invention can integrate internal and external resources and maximize the use of intelligent guidance replenishment, transportation and production.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

FIG1 is a schematic diagram of the main process of supply chain intelligent planning according to the present invention;

FIG2 is a schematic diagram of the association between the product structure/supply chain network and constraints in this embodiment;

FIG3 is a schematic diagram of LTR calculation of available dates of each node in this embodiment;

FIG. 4 is a schematic diagram showing the main purpose of RTLBackwards in this embodiment.

DETAILED DESCRIPTION

The purpose of the present invention is to provide a supply chain intelligent planning method, system, electronic device and storage medium, which can integrate internal and external resources and maximize the use of intelligent guidance for replenishment, transportation and production.

In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

As shown in FIG1 , the present invention provides a supply chain intelligent planning method, comprising:

Step 100: Determine the demand sorting according to the order;

Step 200: Using the product structure model and allocation constraints, formulate demand plans one by one according to the demand ranking to obtain the supply chain product production planning scheme; the product structure model is composed of the MLBOM multi-layer product structure; the allocation constraints are based on the constraint quantity based on the time unit, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraint includes the daily output that the production line can provide, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate per day. Among them, each Supply represents a purchase order, a planned purchase order, a transfer order, a planned transfer order, a production order, a planned production order, etc. in the supply chain, and the supply will be received soon; each unit Supply is the corresponding unit in the supply document, such as pieces, kilograms, etc.

Based on the above technical solution, the following embodiments are provided.

In the product structure model, the most basic is the single-layer BOM structure, with BOM as the parent and COM as the child. In the supply chain field, it flows from a certain node (child node) to another node (parent node), such as material B is manufactured in warehouse L01 of factory 01, and material A is manufactured and stored in L02 under factory 01. Then material B, factory 02 and warehouse L01 form node A, and factory and warehouse L02 form another node. In this embodiment, for ease of explanation, A and B are used to represent nodes.

As shown in Table 1, PER represents how many units of child nodes can be transformed into one unit of parent node, and LT is LeadTime, which is the time required for a child node to be transformed into a parent node without considering constraints.

Table 1 Single-layer BOM structure

BOM COM PER LT A B 1 1 A C 2 1 B D 2 1 B E 1 1 C F 2 1 A1 C1 1 2

Based on the concept of single-layer BOM structure, the MLBOM multi-layer product structure is further constructed, that is, the parent-child relationship can be transformed into a supply chain network. In this embodiment, only the BOM bill of materials is used for illustration. As shown in Table 2, taking a certain node as the starting point, you only need to look at its descendants. Then the starting point is the ROOT root, COM is the child node, and BOM is the parent node of the node. The node as the starting point has no parent node. ROOTLT is the total lead time from a certain node to ROOT, ROOTPER is how many units of nodes can be converted into one unit of ROOT, and the node without child nodes is Leaf.

Table 2 MLBOM multi-layer product structure

Constraint model: When determining the dates on which supply and demand will be available, constraints such as materials and capacity need to be considered simultaneously.

Capacity constraints can be fixed or variable factors that represent production capacity, transportation or supplier capacity. The fundamental constraint is still created at the supply source. Constraints are based on rate, which is the amount of constraint available per time period. For example, the limit on the production process can be the number of production hours that the production line runs. A supply source can have multiple constraints, for example, suppliers are subject to capacity constraints and transportation capacity constraints; a production line has both equipment constraints and personnel constraints; a product may also be subject to multiple layers of constraints, with this layer being subject to capacity constraints and also to the capacity constraints of the semi-finished products in the lower layer. Multiple constraints on the same supply source should consider bottleneck constraints or replacement constraints. For example, the production capacity constraint is 50, the transportation constraint is 40, and when considering its throughput, it should be 40. A product can be produced on multiple production lines.

The multiple constraints of the product structure should be superimposed to see the final impact. Therefore, the constraint model should include the following aspects: the number of constraints based on time units, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit of Supply, and the limit of the constraint, such as the remaining execution quantity of the contract; the capacity constraint is stored in the Constraint table as an operational data model, which can represent: the output that the production line can provide every day, the fixed number of SKUs associated with the supply source, and the number of hours the machine can operate every day. Among them, the constraint Constraint is shown in Table 3, the control group ConstraintGroup is shown in Table 4, the constraint availability ConstrainRate is shown in Table 5, the constraint type ConstraintType is shown in Table 6, and the source constraint SourceConstrain is shown in Table 7.

Table 3 Constraint

Table 4 Control Group ConstraintGroup

FieldName Description DataType Key Group Unique identifier String Yes Description describe String

Table 5 ConstrainRate

Table 6 Constraint Type ConstraintType

Table 7 Source Constrain

The SourceConstraint table is used to associate constraints with the component sources they constrain. Each SourceConstraint record references a MaterialSource record that identifies the source of the supply and a Constraint record that identifies the effective constraint. The constraint referenced in the Constraint record then constrains the component source referenced in the MaterialSource record. The SourceConstraint table also contains three fields that affect the amount of constraint required for the consumables from the associated material source. The SourceConstraint. FixedConstraintFactor field defines the fixed constraint amount to be consumed for each supply from the referenced component source, Before is the preceding fixed consumption, and After is the subsequent consumption. And SourceConstraint. ConstraintFactor defines the variable constraint amount to be consumed for each unit in that supply.

If the fixed or variable constraint factors applied to a part source change over time, then SourceConstraint records can be time-phased. For example, if a manufacturing process improves over time, then it may be necessary to reduce the constraint factors. Each SourceConstraint record has an EffectiveIn field that specifies the date on which the fixed and variable constraint factors on the record are effective. These factors are considered effective until the next EffectiveIn or EffectiveOut. SourceConstraint for a material source and constraint combination. If the constraint factors for a given part source do not change over time, then only one SourceConstraint record is required for a given constraint and part source combination (whose EffectiveIn can be set to the default value of "Past", which translates to the current date). However, if the constraint factors change over time, then two or more SourceConstraint records are required to define the time-phased factors for a constraint and part source combination, as shown in the following figure (assuming one constraint factor is effective until November 29, 2016, and a smaller constraint factor is effective from November 30, 2016 onwards). Constraint groups define alternative constraints that can be applied to a material source. Of course, multiple constraints can also be applied to the same material source.

The product structure/supply chain network and constraint associations are further described.

As shown in Figure 2, Part A, B, C... can be considered as nodes of the supply chain network. When a material source corresponds to multiple constraints, if they are in the same group, it is a substitutable relationship. If they are not in the same group or the group is empty, it is a combination relationship, which means that multiple combination relationship Constraints need to complete the same amount of materials in the same period. That is, the minimum value of Rate/Factor can be used. Therefore, the supply model Supply can be obtained based on the above relationship. As shown in Table 8, there is Supply coded as the planned arrival line ScheduleReciept, Source Make is a production order, and Procurement is a purchase order. Source OH represents the existing inventory.

Table 8 Model parameters

Supply Item FinishDate EffectiveQty Source Material AvailableDate MO1 7 5 Make A 7 PAST 10 OH A PAST MO2 7 15 Make B 7 PAST 80 OH F PO1 1 10 120 Procurement F 10 PAST 100 OH D PAST PO2 1 11 40 Procurement D 11 PO2 2 14 60 Procurement D 14 PAST 80 OH E PAST PO3 1 10 40 Procurement E 10 PO3 2 15 80 Procurement E 15 100 OH C1 116 OH C

Then, a demand model is constructed based on the order. In the independent demand Demand Order, as shown in Tables 9 and 10, DemandType = Sales represents a sales order, and Forecast represents a forecast.

Table 9 Independent Demand Order

Table 10 Example

Then, Allocation is allocated. Allocation is the production order line item, which is the demand data. As shown in Table 11-Table 12, Allocation is the relevant demand or reservation corresponding to the created work order.

Table 11 Allocation

Table 12 Example

The output data is calculated through the above models. PlannedOrder is generated by the LTR process to obtain the earliest available date of the node, as shown in Table 13.

Table 13 PlannedOrder

FieldName Description DataType Key PlannedOrder Plan number String Y Ttem Planned order line items Material Driven material Site Factories, distribution centers, etc. RequestQty Required quantity EMST Earliest time for materials to be complete Int EPST Earliest planned document start time String EPFT Earliest planned document completion date String

The planned order CTPPlannedOrder is generated by the RTL Backwards process and is used to calculate the planned document based on the achievable plan, as shown in Table 14.

Table 14 Planned Order CTPPlannedOrder

FieldName Description DataType Key PlannedOrder Plan number String Y Item Planned order line items Material Driven material Site Factories, distribution centers, etc. RequestQty Required quantity EMST Earliest time for materials to be complete Int EPST Earliest planned document start time String EPFT Earliest planned document completion date String PST DueDate

The related requirement PlannedAllocation is generated by PlannedOrder, as shown in Table 15.

Table 15 Planned Allocation

FieldName Description DataType Key PlannedOrder Plan number String Y Item Planned order line items BOM Assembly Part Number Material Related demand materials Site Factories, distribution centers, etc. RequestQty Required quantity

Among them, CTPPlannedAllocation is generated by TPPlannedOrder, as shown in Table 16.

Table 16 CTPPlannedAllocation

FieldName Description DataType Key PlannedOrder Plan number String Y Item Planned order line items BOM Assembly Part Number Material Related demand materials Site Factories, distribution centers, etc. RequestQty Required quantity DueDate Requirement Date

The demand-supply matching is shown in Table 17. The consumption of constraints is shown in Table 18.

Table 17 DemandSupplyPegging

Table 18 Constraint consumption ConstraintUsed

Finally, the process logic and algorithm of this embodiment are as follows.

Determine the effective order sequence: Use DemandOrder to find the demands with Effective Qty>0, sort them in ascending order by Sequence, and process the demands one by one.

LTR calculates the available date of each node: EMST (Earliest Material Start Date) is the earliest material start date, that is, the earliest material set date. EPST (Earliest Planned Start Date) is the earliest possible start date after considering constraints. LT (Lead Time) refers to the time to complete a task. Here EPST+LT = EPFT (Earliest Planned Order Finished Date) is the earliest possible completion date. This can be used as the earliest available date for the supply chain node/material. As shown in Figure 3. Starting from the bottom layer, leaf = Y, the Level is calculated from large to small.

Match product structure and supply: Match the available supply and product structure and sort them in ascending order by Available Date, as shown in Table 19.

Table 19 Example

Different BOMs calculate the EMST of the BOM separately (in parallel).

Complete the available date data, as shown in Table 20: 1. Find the Available Date set B{6, 10, 11, 14, 15}, C{6, 10}, BD{6, 11, 14}, BE{6, 10, 15} in the same BOM. 2. Insert the BE complement set B-BE{11, 14}, BD complement set {10, 15} corresponding to the COM Available Date in the same BOM. 3. Copy the most recent and earlier Available information for other information, and set Available Qty = 0.

Table 20 Completion Example

The EMST/set calculations are shown in Table 21.

1. There is a series of data, where the date column Date should be the available date corresponding to the supply. In this case, it is AvailableDate, and the number column Qty is Available Qty in this case.

2.SumQty in this example is the sum of Qty (Available Qty)/PER in the same group of SumBOM (BOM+COM) in the order of data arrangement (Date/AvailableDate ascending). LSumQty is SumQty–Qty/PER, in this example it is LSummBOM Qty.

3.MinS in this example is MinSumBOM, which is the minimum SumQty value of the same date items in different BOM+COM data under the same BOM. MinLS is the MinS of the previous data, in this example, the same column MinLSBOM. SquareQty is the complete set quantity, which is equal to MinS-MinLS.

4.Allocated Qty = Squared Qty * PER, the number of COM allocated

5. The Available Date corresponding to Squared Qty>0 is the EMST of BOM.

Table 21 PlannedOrderAnalysis

Generate a planned order and calculate EPST as shown in Table 22. The corresponding material is BOM and Qty is Squared Qty.

Table 22 Example

PlannedOrder Item Material Qty Source EMST Constrain PLO1 B 40 6 4 PLO2 B 10 10 4 PLO3 B 10 11 4 PLO4 B 40 15 4 PLO5 C 40 6 3 PLO6 C 60 10 3

PlannedAllocation is the related demand of the planned order, as shown in Table 23. The corresponding Material is COM, Quantity is Allocated Qty, and EDueDate = EMST.

Table 23 Example

The allocation of plan-related demand PlannedDemandSupply is shown in Table 24.

Table 24 Example

The process of calculating ECST is as follows: Planned Order sorting: Calculation is required when the corresponding Type in SourceConstraint is Constrained. Planned Order associates Constraint with Material. Planned Orders corresponding to the same Constraint are sorted in ascending order by EMST. Different Constraints are processed separately, and ECST is calculated for each Planned Order of the same Constraint in order.

As shown in Table 25, the corresponding Planned Orders in Constraints 4 and 3 are processed simultaneously, and the Planned Orders in the same Constraint are processed one by one in ascending order of EMST. For example, in Constraint 4, PLO1 is processed first and then PLO2.

Table 25 Example

PlannedOrder Material EMST Constrain PLO1 B 6 4 PLO2 B 10 4 PLO3 B 11 4 PLO4 B 15 4 PLO5 C 6 3 PLO6 C 10 3

Calculate ECST with the data of Available Date>=EMST in Material-associated Constraint. Algorithm for calculating available Constraints. The data model of Constraints used by Scheduled Receipts (SRs) and Planned Orders is SupplyConstraint, as shown in Table 26. SumConstraintBYDate is the summary of AllocatedRate for the same Constraint and the same available date. Available Rate=Rate–AllocatedRate. Available volume/rate. AllocatedRate=Effective Qty*Factor. If the corresponding Constraint has FixedFactor. AllocatedRate=Effective Qty*Factor+FixedFactor. FixFactor represents fixed consumption, such as setup, waiting, clearing, etc.

Table 26 SupplyConstraint

Calculate EPST based on the data in the Material-associated Constraint where Available Date >= EMST.

Allocation algorithm:

1. Determine the required quantity, in this case RequestQty.

2. Determine the sorting rules for the objects to be assigned.

3. AvailableQty is the available supply quantity, and SUMQty is the cumulative AvailableQty quantity according to the sorting rule (calculated earliest, so AvailableDate is used in ascending order in this stage).

4.SUMLQTY is the SUMQty accumulated to the previous data, which is equivalent to SUMQty-AvailableQty of this data.

5.Min{RequestQty,SUMQty}＝AllocatedSUMQty, where RequestQty and SUMQty need to be converted to the same units, for example, if there is unit conversion, the per unit usage of BOM, etc.

6.AllocatedSUMLQty: AllocatedSUMQty of the previous number.

7.AllocatedQty=AllocatedSUMQty-AllocatedSUMLQty.

8. Data with AllocatedQty>0 is valid data, and AllocatedQty is the number of objects allocated.

Use the allocation algorithm to calculate the allocation of Constraint to Planned Order. In addition:

1. If the objects in the plan are all in integer units such as pieces/units, then the calculation process should be rounded to the left. For example, if Rate/Factor is the number of units that can be produced, it should be rounded to the left. That is, 5.1 and 5.7 are both 5. For other decimals that are allowed, take normal decimals.

2. When the Constraint corresponding to the material is Load or Ignore, there is no need to calculate the Constraint allocation through this algorithm. Directly generate CTPPlanned Order, EMST = EPST. EMFPQty = EMQty. EPFT = EPST + LTIgnore's Constraint is empty, and Load's Constraint is filled in.

3. When calculating BOM, sometimes it is necessary to consider the loss rate Yield. Then EMQty = SquaredQty*1/(1+Yield). An example is shown in Table 27.

Table 27 Example

From this we can see that PLO1 has 11 units that can start on Date 6, and 29 units that start on Date 7. PLO5 has 40 units that start on Date 6.

The resulting planning result is: When the Constraint is considered, the PlannedOrder data will be updated, and the EPST earliest start time, earliest completion time, and start quantity EPSTQty will be calculated. PlannedOrder will be used as Supply, EPFT as the AvailableDate of the supply, and EPFTQty as the AvailableQty.

Continue the Planned Order calculation until the end, and the results are shown in Table 28.

Table 28 Planned Order

PlannedOrder Item Material Qty Source EMST Constrain PLO1 B 40 6 4 PLO2 B 10 10 4 PLO3 B 10 11 4 PLO4 B 40 15 4 PLO5 C 40 6 3 PLO6 C 60 10 3

At this time, whether the Material is ROOT, or whether the Level is 0. If it is 0, then start calculating the delivery time. Otherwise, continue to calculate the previous level. From COM=B, C, it can be known that A will be calculated. Then go through: matching product structure and supply, different BOMs are calculated separately (in parallel) and ECST calculation.

As another example, material A can be consumed in both Constraint 1 and Constraint 2.

Replace the matching algorithm of Constraint: Assume that material A01 corresponds to Constraint1,2,3. The priority is 1,2,3. In SourceConstraint, AllocateIntervals is set to 3, and the currently applied Calendar is day. Then PlannedOrder will match the current interval (today) when calculating EPST, assuming that today's Date is 6. The results are shown in Table 29.

Table 29 Results

RequestQty Material Constain AvailableDate AvailableRate Priority Group 100 A01 1 6 1 1 100 A01 1 7 1 1 100 A01 1 8 1 1 100 A01 2 6 2 1 100 A01 2 7 2 1 100 A01 2 8 2 1 100 A01 3 6 3 1 100 A01 3 7 3 1 100 A01 3 8 3 1

When AllocateIntevals is set to 2, the delivery date determines that the planned order PST of a certain A01 is Date 16 (this process is RTLBackward). The matching results are shown in Table 30.

Table 30 Matching results

RequestQty Material Constain AvailableDate AvailableRate Priority Group 100 A01 1 16 1 1 100 A01 1 15 1 1 100 A01 2 16 2 1 100 A01 2 15 2 1 100 A01 3 16 3 1 100 A01 3 15 3 1

Back to this example, the AllocateIntervals of Constraint1,2 is 2. Applying the matching algorithm, PlannedOrder is obtained, as shown in Table 31.

Table 31 PlannedOrder

A is the required material. So far, the earliest available date and quantity of A have been obtained, as shown in Table 32.

Table 32 Example

Develop a delivery plan and determine the sorting rules for the assigned objects: 1. The supplies bound to this demand, that is, the supplies that exist in DemandSupply and correspond to the demand, are sorted in descending order of Available Date and Available Qty. 2. For Planned Order, its AvailableDate = EPFT. 3. When AvailableDate < = RequestDate, the non-bound supplies are sorted in descending order of AvailableDate.

also:

1. Although some existing inventory and planned arrivals are bound to other demands, if the supply quantity < the allocated quantity, i.e. EffectiveQty < AllocatedQty, then this supply is equivalent to having some parts that are not bound, and EffectiveQty – AllocatedQty is used as AvailableQty. 2. When the bound demand is invalid (such as order cancellation), this supply is also unbound supply. 3. When AvailableDate > RequestDate, unbound supplies are sorted in ascending order by AvailableDate.

Calculate DueDate using the allocation algorithm: DueDate is the planned demand date of the corresponding allocated Supply. If the Supply corresponds to the Planned Order, then the corresponding date is DueDate, and this Date-LT will be used as the demand date for the Constraint to start backwards.

1. AllocatedQty=0 is invalid data.

2.LastDate is the maximum value of AvailableDate when AllocatedQty>0.

3. If the requirements are met by a planned document, a planned document will be generated later. The new planned document will maintain an association with the previous planned document. The order quantity will be updated with the allocated quantity.

4. When LastDate>RequestDate, DueDate＝AvailableDate. The specific process is:

a) The quantities of the planned documents are based on SumByDateQty, PST = EPST, PFT = EFT. b) If the order cannot be delivered ‘Partially’, the latest DueDate is the delivery date and the quantity is the RequestQty quantity. c) Otherwise, SumByDateQty is the aggregate quantity of the same DueDate, Type. This quantity is the final delivery plan quantity. d) SumByDateQty is used as the driving quantity of the planned document.

5. When LastDate<=RequesDate, DueDate=LastDate

The DueDate of the planned document to be generated is deduced backward and regenerated, that is, the calculation needs to be reallocated. Filter Constraint by AvailableDate <= DueDate-LT and sort in descending order. Assume that the demand SO1, RequestQty is 100, and RequestDate is Date 14. The result is shown in Table 33.

Table 33 Results

Example 10: Delivery plan: Assume that the demand is SO1. Request Qty is 30, RequestDate is Date 14. And there is no binding relationship between A's OH and MO1, then the OH inventory and MO1 corresponding to A in the result are not allocated, they can be allocated to priority orders or new orders. RTLBackwards generates a planning document and allocates resources, as shown in Figure 4. The main purpose of RTLBackwards is: 1. Make the overall supply chain smoother; 2. Release the materials and resources occupied earlier so that subsequent demand can be met faster.

Main process: 1. Clear the planned output data except PlannedOrder in LTR. Such as PlannedAllocation, PlannedDemandSupply. 2. Sort the supply in descending order based on DueDate. 3. In "Calculate DueDate using allocation algorithm", you can already know how to allocate supply and generate planning documents. 4. Generate related demand from planning documents. 5. Continue to step 3 until Leaf.

Generate a planning document: Calculate Constrain allocation according to the allocation algorithm: 1. When the Type corresponding to the Constraint is LoadOnly: a) No need to calculate through the Constraint allocation. DueDate–LT＝PST, directly generate CTPPlannedOrder. b) Generate data in ConstraintUsed, AllocatedRate＝CTPPlannedOrder. Qty*Factor. 2. When the Type corresponding to the Constraint is Unconstrained, it is ignored. 3. When the Type corresponding to the Constraint is Constrained, it needs to be calculated using the allocation algorithm. The calculation results are as follows based on the assumption of Example 10, and all Constraints need to be calculated.

DueDate-LT is the date to be picked up and reversed to Available Date. At this time, CTPPlannedOrder is generated by constraint allocation: Constraint consumption in different periods generates different CTPPlannedOrder. EPFT takes the AvailableDate of the calculated part in the delivery plan. CTP-related demand CTPAllocation: CTPPlannedOrder generates related demand according to the BOM model. Further allocate resources: demand and supply match DemandSupplyPegging, then constraint consumption ConstraintUsed, and loop the Backwards logic until no more demand is generated.

CTPPlannedAllocation is the demand utilization allocation algorithm that continues to go backwards. MO2 is not allocated and can be used for subsequent orders. The results are shown in Table 34.

Table 34 Results

CTPPlannedOrder Item Material Quantity PerQty DueDate Supply AvailableDate AvailableQty EPST AllocatedQty SumByDateQty CTPPLO1 1 B 30 1 12 PLO2 12 10 11 10 30 CTPPLO1 1 B 30 1 12 PLO3 12 10 11 10 30 CTPPLO1 1 B 30 1 12 PLO1 8 29 7 10 30 CTPPLO1 1 B 30 1 12 PL01 7 11 7 0 CTPPLO1 1 B 30 1 12 MO2 7 13 0 CTPPLO1 2 C 60 2 12 PLO6 12 60 11 60 60 CTPPLO1 2 C 60 2 12 PLO5 7 40 6 0 0 CTPPLO1 2 C 60 2 12 CTPPLO1 2 C 60 2 12 CTPPLO1 2 C 60 2 12

Among them, B and C are obtained by inverting ConstrainRate allocation with DueDate-LT of 30 and 60 as the date node, as shown in Table 35.

Table 35 Example

Supply Material Planned Constrain AllocatedQty AllocatedRate AvailableDate Factor CTPPLO1 A Y 1 30 3 13 0.1 CTPPLO2 B Y 4 30 3.6 11 0.12 CTPPLO3 C Y 3 60 3 11 0.05

Therefore, the generated CTPPlannedOrder is shown in Table 36.

Table 36 CTPPlannedOrder

Then, ConstrainUsed is obtained as shown in Table 37, DemandSupplyPegging is shown in Table 38, and CTPPlannedAllocation is shown in Table 39.

Table 37 ConstrainUsed

Supply Material Planned Constrain AllocatedQty AllocatedRate AvailableDate Factor CTPPLO1 A Y 1 30 3 13 0.1 CTPPLO2 B Y 4 30 3.6 11 0.12 CTPPLO3 C Y 3 60 3 11 0.05

Table 38 DemandSupplyPegging

SO1 1 A SO1 1 A CTPPLO1 A 30 SO1 1 A CTPPLO1 1 B CTPPLO2 B 30 SO1 1 A CTPPLO1 2 C CTPPLO3 C 60

Table 39 CTPPlannedAllocation

CTPPlannedOrder Item Material Quantity PerQty DueDate CTPPLO1 1 B 30 1 13 CTPPLO1 2 C 60 2 13 CTPPLO2 1 D 60 2 11 CTPPLO2 2 E 30 1 11 CTPPLO3 1 F 120 2 11

Continue to use CTPPlannedAllocation as the demand and DueDate to allocate the supply, as shown in Table 40.

Table 40 Example

This allocation result can no longer generate corresponding demand, and DemandSupplyPegging is generated to carry out the next order/demand plan. Finally, the resource utilization rate is continuously improved, and the process is:

1. Determine the planning interval: a) Plan the workflow in the system and set the time for each corresponding plan. b) The interval between this plan and the next plan is the time interval that needs to be considered.

2. Determine the idle resources within the interval: a) Traverse all Constraint data according to the interval in step 1 b). b) Constraint data with Rate–∑AllocatedRate>0 is valid data.

3. Fill resources: a) Traverse the CTPPlannedOrder or SR whose EPST falls within this interval. b) Data with EPST < PST is valid. c) Sort in ascending order according to PST and process them one by one. d) Clear the relevant materials and resource allocation (single layer) corresponding to the current CTPPlannedOrder or SR, that is, FatherDemand is the CTPPlannedOrder or SR. e) Recalculate the points using the Forward algorithm until Rate–∑AllocatedRate＝0 or the relevant CTPPlannedOrder or SR is processed.

This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the core idea of the present invention. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims (8)

1. A supply chain intelligent planning method, characterized by comprising: Determine the order of demand based on the order; By using the product structure model and allocation constraints, demand plans are formulated one by one according to the demand ranking to obtain the supply chain product production planning scheme; the product structure model is composed of an MLBOM multi-layer product structure; the allocation constraints are based on the constraint quantity based on the time unit, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraints include the output that the production line can provide per day, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate per day. 2. The supply chain intelligent planning method according to claim 1, characterized in that the process of formulating the demand plan is: In LTR Forward, the availability date of each production node is determined, and a delivery plan is formulated based on each availability date. RTL Backward uses the delivery date as a benchmark and uses the constraint model to allocate resources layer by layer to generate planned orders. 3. The supply chain intelligent planning method according to claim 1 is characterized in that the MLBOM multi-layer product structure is a supply chain network transformed based on the single-layer BOM structure; in the supply chain network, the starting point is the ROOT root, COM is the child node, BOM is the parent node of the node, the node as the starting point has no parent node, ROOTLT is the total lead time from a certain node to ROOT, ROOTPER is how many units of nodes can be converted into one unit of ROOT, and the node without child nodes is Leaf. 4. The supply chain intelligent planning method according to claim 3 is characterized in that the single-layer BOM structure includes a parent node BOM and a child node COM; in the supply process, when flowing from any node to another node, any node is the child node COM, and the other node is the parent node BOM. 5. The supply chain intelligent planning method according to claim 1 is characterized in that, in the MLBOM multi-layer product structure, when a material source corresponds to multiple constraints, if they are in the same group, they are substitutable relationships; if they are not in the same group or the group is empty, they are combination relationships, which means that multiple combination relationships need to complete the same amount of materials in the same period, and the minimum value of Rate/Factor is taken. 6. A supply chain intelligent planning system, characterized by comprising: A sorting determination unit, used to determine the demand sorting according to the order; The supply planning unit is used to use the constraint model to formulate demand plans one by one according to the demand ranking to obtain the supply chain product production planning scheme; the constraint model is based on the constraint quantity based on the time unit, the fixed constraint consumption quantity of each Supply, the consumption quantity of each unit Supply, the constraint limit and the capacity constraint; the capacity constraint includes the output that the production line can provide every day, the fixed number of SKUs associated with the supply source and the number of hours the machine can operate every day; each Supply represents the flow process of goods, services or resources provided by a certain link in the supply chain to the next link; each unit Supply represents the specific quantity of goods, services or resources provided by a certain link in the supply chain to the next link. 7. An electronic device, characterized in that it comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the supply chain intelligent planning method according to any one of claims 1-5. 8. A computer-readable storage medium, characterized in that it stores a computer program, and when the computer program is executed by a processor, it implements the supply chain intelligent planning method as described in any one of claims 1 to 5.