KR102591580B1 - Construction plan information analysis-based construction bolt demand forecasting and inventory management method, device and system - Google Patents

Abstract

One embodiment of the present invention predicts future demand for each type of construction bolt based on existing construction plan information, new construction plan information, and changes in inventory by type of construction bolt, and orders construction bolts based on the predicted demand to meet future needs. It relates to a demand forecasting and inventory management method, device, and system for construction bolts based on analysis of construction plan information that secures inventory in advance but controls the order quantity to prevent malicious inventory from occurring.

Description

Demand forecasting and inventory management method, device and system for construction bolts based on construction plan information analysis {CONSTRUCTION PLAN INFORMATION ANALYSIS-BASED CONSTRUCTION BOLT DEMAND FORECASTING AND INVENTORY MANAGEMENT METHOD, DEVICE AND SYSTEM}

The examples below predict future demand for each type of construction bolt based on existing construction plan information, new construction plan information, and changes in inventory by type of construction bolt, and order construction bolts based on the predicted demand to prepare inventory needed in the future in advance. It relates to a demand forecasting and inventory management method, device, and system for construction bolts based on analysis of construction plan information that secures the order volume while controlling the order quantity to prevent malicious inventory from occurring.

Construction bolts of different specifications are used in various construction projects such as plants, marine plants, residential building construction, commercial building construction, factory construction, and civil engineering construction.

The specifications/dimensions of these construction bolts are very diverse, so it is very difficult for construction companies to directly manage the inventory of these construction bolts. In the construction planning stage, the required quantity is ordered and received from the construction bolt manufacturer or distributor. is managing.

Meanwhile, construction bolt manufacturing/distribution companies manage the inventory of numerous construction bolts in various ways, but it is difficult to accurately predict the demand for each type, and when the inventory of some construction bolts is excessive, they remain as toxic inventory and are then discarded due to corrosion, etc. etc., causing losses.

Although it is possible to predict some demand based on conventional artificial intelligence demand forecasting technology, it is very difficult to accurately predict all demand for each of the many different types of construction bolts, and data is insufficient, especially for types that are used only in small quantities. Therefore, there is a high probability that demand will be predicted incorrectly.

KR 10-0928621 B KR 10-2016-0085373 A

The problem to be solved by an embodiment of the present invention is based on existing construction plan information, new construction plan information, and inventory changes by type of construction bolts in order to overcome the limitations of the conventional construction bolt (part) demand forecasting technology as described above. By predicting the future demand for each type of construction bolt, and calculating the inventory/demand for each type of construction bolt based on a clustering algorithm, it is possible to accurately predict demand for types of construction bolts for which there is insufficient existing data, and order construction bolts based on this. The purpose is to provide a demand forecasting and inventory management method, device, and system for construction bolts based on analysis of construction plan information that can secure the necessary inventory in advance and control the order quantity to prevent malicious inventory from occurring.

According to one embodiment, a method for predicting demand and managing inventory of construction bolts based on analysis of construction plan information performed by an apparatus includes: receiving new construction plan information; Step of receiving order information for construction bolts; Recording inventory change information of the construction bolts; Predicting the demand for the construction bolts based on new construction plan information, existing construction plan information, and inventory change information; and ordering construction bolts based on the predicted demand for construction bolts. It provides a demand forecasting and inventory management method for construction bolts based on analysis of construction plan information, including a method.

In addition, the construction bolts are: stored inside a predetermined warehouse, where a plurality of shelves are arranged inside the warehouse, a plurality of trays are placed on the shelves, and the construction bolts are accommodated in different trays for each type, Moving construction bolts within the warehouse; further comprising: recording the inventory change information of the construction bolts: measuring the weight by type of the construction bolts from a weight sensor installed on a tray accommodating the construction bolts by type; measuring; Recording the weight of each type of construction bolt by time; And calculating a change in inventory by time for each type of the construction bolt, based on the weight by time for each type of the construction bolt; including, the step of moving the construction bolts inside the warehouse where the construction bolts are stored: ', generating a first matrix with the types of construction bolts as 'rows' and the inventory change by time period as 'values'; Based on the first matrix, clustering the types of construction bolts into first clusters using K-Means clustering for the amount of change in inventory by time period for each type of construction bolts; Distributing and designating storage locations for each type of construction bolts included in each first cluster to different shelves; And it may include the step of controlling a predetermined AGV to transfer the tray, depending on the storage location for each type of construction bolt.

And, the step of predicting the demand for construction bolts includes: calculating a first partial demand based on the new construction plan information and the existing construction plan information; calculating a second partial demand based on the new construction plan information and the existing construction plan information; calculating a third partial demand based on the inventory change information; And calculating the demand for construction bolts by reflecting a predetermined weight for each of the first to third partial demands and adding them up, wherein the step of calculating the first partial demand includes: the new construction plan Creating a second matrix with the construction price included in the information and the existing construction plan information, the correlation by construction type, the construction start month, and the construction period as 'columns' and the unique number of the construction as 'rows'; standardizing the values of the second matrix by dividing them by the average value for each column; Based on the standardized second matrix, clustering the construction projects into second clusters using K-Means clustering for the new construction plan information and the existing construction plan information; Among the second clusters, designating a third cluster containing new construction; And a step of designating the average value of the 'order quantity by type of construction bolts included in the third cluster' as the first partial demand based on 'order information for construction bolts of construction companies included in the third cluster', The step of calculating the second partial demand includes: document clustering the construction plans included in the new construction plan information and the existing construction plan information into fourth clusters using K-Means clustering; Document clustering the detailed schedules included in the new construction plan information and the existing construction plan information into fifth clusters using K-Means clustering; Designating the average value of the 'order quantity by type of construction bolts included in the fourth cluster' as the 2-1 partial demand based on 'order receipt information for construction bolts of construction companies included in the fourth cluster'; Designating the average value of the 'order quantity by type of construction bolts included in the 5th cluster' as the 2-2 partial demand based on 'order receipt information for construction bolts of construction companies included in the 5th cluster'; And a step of designating the average value of the 2-1 partial demand quantity and the 2-2 partial demand quantity as the second partial demand quantity; calculating the third partial demand quantity includes: calculating the third partial demand quantity from the inventory change information, Creating a third matrix with '(Month)' as 'column', the type of construction bolt as 'row', and the monthly inventory for each type of construction bolt as 'value'; Based on the third matrix, clustering the types of construction bolts into sixth clusters using K-Means clustering for monthly inventory by type of construction bolts; And, based on the monthly inventory by type of construction bolts included in the sixth cluster, designating the average value of 'monthly inventory by type of construction bolts included in the sixth cluster' as the third partial demand.

In addition, the step of calculating the first partial demand includes: selecting a reference target construction based on the existing construction plan information; Extracting the existing construction start month from the existing construction plan information corresponding to the reference target construction; extracting a new construction start month from the new construction plan information; Calculating a difference between the existing construction start month and the new construction start month; calculating a first weighting ratio by dividing the difference by 12; and recalculating the first partial demand by subtracting a value obtained by multiplying the first partial demand by the first weighting rate from the first partial demand. The step of calculating the second partial demand includes: extracting an existing construction plan and an existing detailed schedule from existing construction plan information corresponding to the reference target construction; Extracting a new construction plan and a new detailed schedule from the new construction plan information; Calculating a first document similarity between the existing construction plan and the new construction plan; Calculating a second document similarity between the existing detailed schedule and the new detailed schedule; Calculating a second weighting rate by adding 0.1 to the first document similarity, and correcting the calculated second weighting rate to 1 if it exceeds 1; Calculating a third weighting ratio by adding 0.1 to the second document similarity, and correcting the calculated third weighting ratio to 1 if it exceeds 1; recalculating the 2-1 partial demand by multiplying the 2-1 partial demand by the second weighting factor; and recalculating the 2-2 partial demand by multiplying the 2-2 partial demand by the third weighting factor.

In addition, the step of predicting the demand for construction bolts includes: calculating a root mean square value for the amount of inventory change by type of construction bolts from the monthly inventory by type of construction bolts; For each type of construction bolt, calculating a first frequency at which the inventory change amount for each type of construction bolt exceeds the root mean square value; Calculating a change in inventory by type of construction bolts over the past year from the monthly inventory by type of construction bolts; Calculating a second frequency at which the change in inventory for each type of construction bolt over the past year exceeds the root mean square value; And when the second frequency is greater than or equal to the first frequency: in the demand quantity for construction bolts, a step of correcting the order quantity for each type of construction bolt with the root mean square value.

The device according to one embodiment may be combined with hardware and controlled by a computer program stored in a medium to execute any one of the above-described methods.

According to one embodiment, future demand for each type of construction bolt can be predicted based on existing construction plan information, new construction plan information, and inventory changes by type of construction bolt.

In addition, by calculating inventory/demand for each type of construction bolt based on a clustering algorithm, it is possible to accurately predict demand for types of construction bolts for which there is insufficient existing data.

In addition, by ordering construction bolts based on the predicted demand for each type of construction bolt, it is possible to secure inventory needed in the future in advance.

In addition, by dispersing the shelves (trays) where construction bolts are shipped inside the warehouse storing construction bolts, the time required for shipping can be shortened by preventing movement lines between AGVs from overlapping when shipping construction bolts.

In addition, the order quantity can be adjusted to prevent malicious inventory by type of construction bolt.

Figure 1 is a flowchart of a method for predicting demand and managing inventory for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.
Figure 2 is a flowchart showing the steps of recording inventory change information of construction bolts in the demand forecasting and inventory management method for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.
Figure 3 is a flowchart showing the steps of moving construction bolts inside a warehouse in the demand forecasting and inventory management method for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.
Figure 4 is a flow chart showing the steps of predicting the demand for construction bolts in the demand forecasting and inventory management method for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.
Figure 5 is a flowchart showing the step of calculating the first partial demand of the demand forecasting and inventory management method for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.
Figure 6 is a flowchart showing the step of calculating the second partial demand in the demand forecasting and inventory management method for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.
Figure 7 is a flowchart showing the step of calculating the third part of the demand forecasting and inventory management method for construction bolts based on analysis of construction plan information according to an embodiment of the present invention.

Hereinafter, embodiments will be described in detail with reference to the attached drawings. However, various changes can be made to the embodiments, so the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents, or substitutes for the embodiments are included in the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Terms such as first or second may be used to describe various components, but these terms should be interpreted only for the purpose of distinguishing one component from another component. For example, a first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being “connected” to another component, it should be understood that it may be directly connected or connected to the other component, but that other components may exist in between.

The terms used in the examples are for descriptive purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by a person of ordinary skill in the technical field to which the embodiments belong. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless explicitly defined in the present application, should not be interpreted in an ideal or excessively formal sense. No.

In addition, when describing with reference to the accompanying drawings, identical components will be assigned the same reference numerals regardless of the reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiments, if it is determined that detailed descriptions of related known technologies may unnecessarily obscure the gist of the embodiments, the detailed descriptions are omitted.

An artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, it is a system in which machines learn and make decisions on their own. As artificial intelligence systems are used, their recognition rates improve and they can more accurately understand user preferences, and existing rule-based smart systems are gradually being replaced by deep learning-based artificial intelligence systems.

Artificial intelligence technology consists of machine learning and element technologies using machine learning. Machine learning is an algorithmic technology that classifies/learns the characteristics of input data on its own, and elemental technology is a technology that mimics the functions of the human brain such as cognition and judgment by utilizing machine learning algorithms such as deep learning, including linguistic understanding and visual It consists of technical areas such as understanding, reasoning/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology that recognizes and applies/processes human language/characters and includes natural language processing, machine translation, conversation systems, question and answer, and voice recognition/synthesis. Visual understanding is a technology that recognizes and processes objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, spatial understanding, and image improvement. Inferential prediction is a technology that judges information to make logical inferences and predictions, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendations. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification) and knowledge management (data utilization). Motion control is a technology that controls the autonomous driving of vehicles and the movement of robots, and includes motion control (navigation, collision, driving), operation control (behavior control), etc.

Generally, in order to apply machine learning algorithms to real life, learning is performed using a trial and error method due to the nature of the basic methodology of machine learning. In particular, deep learning requires hundreds of thousands of iterations. It is impossible to execute this in an actual physical external environment, so instead, the actual physical external environment is virtually implemented on a computer and learning is performed through simulation.

According to one embodiment, in a method for predicting demand and managing inventory of construction bolts based on analysis of construction plan information performed by an apparatus, the method includes receiving new construction plan information (S100); Step of receiving order information for construction bolts (S200); Recording inventory change information of the construction bolts (S300); Predicting the demand for the construction bolts based on new construction plan information, existing construction plan information, and inventory change information (S400); and ordering construction bolts based on the predicted demand for construction bolts (S500). It provides a demand forecasting and inventory management method for construction bolts based on construction plan information analysis.

The device may be configured as a certain server computer.

In the step of receiving new construction plan information (S100), information related to new construction orders newly awarded by period (monthly) can be input.

The construction plan information entered in this way is continuously accumulated and stored, and can be used as 'existing construction plan information' in the next period (next month). In order to configure an embodiment of the present invention, 'existing construction plan information' accumulated over a predetermined period of time must be provided.

In the step (S200) of receiving order information for construction bolts, the order quantity (order quantity) for each type of construction bolt required for the new construction works is input.

In the step (S300) of recording the inventory change information of the construction bolts, the current inventory is tracked by monitoring the shipment quantity of each type of construction bolts ordered and shipped, and the receipt quantity of each type of construction bolts ordered and received.

In the step (S400) of predicting the demand for construction bolts based on new construction plan information, existing construction plan information, and inventory change information, the construction bolts are used through a predetermined artificial intelligence algorithm based on construction bolts and construction-related data. Predict how many bolts will be needed in the future (how many orders will be received) by type of construction bolt.

In the step (S500) of ordering construction bolts based on the predicted demand for construction bolts, construction bolts are ordered by type based on the difference between the predicted demand for each type of construction bolt and the current inventory for each type of construction bolt. Place an order.

Additionally, the construction bolts may be stored inside a certain warehouse.

Here, a plurality of shelves are arranged inside the warehouse, a plurality of trays are placed on the shelves, and the construction bolts are accommodated in different trays for each type.

For example, 'A' type construction bolts are accommodated in tray 1, type 'B' construction bolts are accommodated in tray 2, and the first and second trays can be placed on the 'ga' shelf. .

Here, the method for predicting demand and managing inventory of construction bolts based on analysis of construction plan information may further include a step (S600) of moving the construction bolts within the warehouse.

The step of recording inventory change information of the construction bolts (S300) includes: measuring the weight of each type of construction bolts from a weight sensor installed on a tray accommodating the construction bolts by type (S310); Recording the weight of each type of construction bolt by time (S320); And based on the weight over time by type of the construction bolt, calculating the amount of change in inventory over time by type of the construction bolt (S330); may include.

Based on the above-described inventory change information based on the arrival and departure quantity, the change in inventory by time for each type of construction bolt may be calculated. However, in one embodiment, in order to reduce errors due to failure to enter the arrival and departure quantity, etc., the type of construction bolt is classified based on the weight of each tray. The amount of inventory change by time can be calculated. The time-dependent inventory change amount for each type of construction bolt calculated based on weight is used in the step of moving the construction bolt inside the warehouse where the construction bolt is stored, and the inventory change information calculated based on the input and output amount is used for demand forecasting, which will be described later. It can be utilized.

The step of moving the construction bolts inside the warehouse where the construction bolts are stored (S600) is: a system in which the time zone is 'column', the type of the construction bolt is 'row', and the inventory change by time zone is 'value'. Step 1: Generating a matrix (S610); Based on the first matrix, clustering the types of construction bolts into first clusters using K-Means clustering for the change in inventory by time period for each type of construction bolts (S620); A step (S630) of distributing and designating storage locations for each type of construction bolt included in each first cluster to different shelves; And it may include a step (S640) of controlling a predetermined AGV to transfer the tray, depending on the storage location for each type of construction bolt.

In the step (S610) of generating a first matrix with the time zone as 'column', the type of construction bolt as 'row', and the inventory change by time zone as 'value', each time zone (e.g., 4 From 12:00 to 5:00 and 5:00 to 6:00), the change in inventory by time period by type of construction bolt is created as a first matrix.

Here, the inventory change amount refers to values such as how many types of construction bolts are shipped and how many are received by time period.

Based on the first matrix, clustering the types of construction bolts into first clusters using K-Means clustering for the change in inventory by time period for each type of construction bolt (S620); In the step (S620), through the inventory change data Cluster types of construction bolts.

For example, A construction bolts and B construction bolts, which have a high shipment volume between 4 and 5 o'clock, are classified into the same first cluster, and C construction bolts and D construction bolts, which have a high shipment volume between 9 and 10 o'clock, are classified into the same first cluster. It can be classified as:

The step (S630) of distributing and designating storage locations for each type of construction bolts included in each first cluster to different shelves distributes storage locations for each type of construction bolts included in a first cluster.

In other words, the A construction bolts, B construction bolts, C construction bolts, D construction bolts, and E construction bolts included in a first cluster may be distributed and arranged on different shelves.

In the step (S640) of controlling a predetermined AGV to transfer the tray according to the storage location for each type of construction bolt, the tray is controlled to be transferred to another shelf based on general AGV-based logistics technology. can do.

And, the step of predicting the demand for construction bolts (S400) includes: calculating a first partial demand based on the new construction plan information and the existing construction plan information (S410); Calculating a second partial demand based on the new construction plan information and the existing construction plan information (S420); Calculating a third partial demand based on the inventory change information (S430); And calculating the demand for construction bolts by reflecting a predetermined weight for each of the first to third partial demands and adding them up (S440).

Here, in the step (S440) of calculating the demand for construction bolts by reflecting a predetermined weight for each of the first to third partial demands, 0.3 for the first partial demand, 0.3 for the second partial demand, By reflecting a weight of 0.4 in the third part demand (so that the sum of each weight is 1), the demand for each type of construction bolt can be calculated.

For example, if the demand for the first part of 'A' construction bolts is calculated as 100, the demand for the second part is 150, and the demand for the third part is 80, the demand for 'A' construction bolts is 100*0.3 + 150*0.3. + 80*0.4 = can be calculated as 107, and if the demand for the first part of 'B' construction bolts is calculated as 500, the demand for the second part is 400, and the demand for the third part is 700, the demand for 'B' construction bolts is can be calculated as 500*0.3 + 400*0.3 + 700*0.4 = 550.

Hereinafter, the specific details of the step of calculating the first partial demand (S410) to the step of calculating the third partial demand (S430) and the step of calculating the demand of construction bolts (S440) will be described.

The step of calculating the first partial demand (S410) is: the construction price included in the new construction plan information and the existing construction plan information, the degree of correlation by construction type, the construction start month, and the construction period as 'columns', and the construction Creating a second matrix with unique numbers as 'rows' (S411); Standardizing the values of the second matrix by dividing them by the average value for each column (S412); Based on the standardized second matrix, clustering the construction projects into second clusters using K-Means clustering for the new construction plan information and the existing construction plan information (S413); Among the second clusters, designating a third cluster containing new construction (S414); and a step (S415) of designating the average value of the 'order quantity by type of construction bolts included in the third cluster' as the first partial demand based on 'order receipt information for construction bolts of construction companies included in the third cluster'; It can be included.

Creating a second matrix with the construction price included in the new construction plan information and the existing construction plan information, the correlation by construction type, the construction start month, and the construction period as 'columns', and the unique number of the construction as 'rows'. In step (S411), the construction price for each construction included in the construction plan information, the correlation by type of construction (plant, marine plant, residential building construction, commercial building construction, factory construction, civil engineering, etc.), construction start month, A second predetermined matrix is generated based on the construction period.

Here, the degree of relevance for each type of construction may refer to the ratio in which each construction is related to the type of construction.

For example, 'A' construction, due to the nature of the construction, has a 70% correlation with 'plant construction', a 50% correlation with 'civil engineering construction', and 0 with 'residential building construction' and 'commercial building construction'. It can have a degree of correlation of %.

This degree of correlation for each type of construction can be entered as a value determined by the manager based on the construction information each time new construction plan information is entered.

At this time, one column of the second matrix can be configured for each type of construction, such as 'plant connection map', 'offshore plant connection map', ..., 'civil construction connection map'.

In the step S412 of standardizing the values of the second matrix by dividing them by the average value of each column, each column of the second matrix may be calculated to have standardized values.

For example, while the 'construction period' has a value of 1 year (12 months) to 3 years (36 months) (values of 12 to 36), the correlation by construction type is a ratio, so it is less than 100% and less than 1. Since it has a value, if it is input into K-Means clustering without standardizing it, the degree of association by type of construction will have a very small influence on the clustering process, so this standardization is done so that the data in each column can have the same/equal influence on clustering. It is advisable to go through the process.

Based on the standardized second matrix, clustering the construction projects into second clusters using K-Means clustering for the new construction plan information and the existing construction plan information (S413); based on the second matrix Cluster construction projects.

For example, 50 construction projects recorded in the construction plan information can be clustered into 8 second clusters.

In the step (S414) of designating a third cluster including new construction among the second clusters, third clusters that are clusters including new construction are designated among the plurality of second clusters.

In the step (S415) of specifying the average value of the 'order quantity by type of construction bolts included in the third cluster' as the first partial demand quantity based on 'order information for construction bolts of construction companies included in the third cluster', Based on the construction bolt order information for the construction projects included in the third cluster, that is, the average value of the shipment quantity (order quantity) for each type of existing construction bolts included in the existing construction plan information, the first partial demand quantity predicted to be ordered in the new construction is calculated. It can be calculated.

The first partial demand for each type of construction bolt calculated in this way can be added to calculate the demand for construction bolts by applying a predetermined weight, for example, a weight of 0.3.

The step of calculating the second partial demand (S420) includes: document clustering the construction plans included in the new construction plan information and the existing construction plan information into fourth clusters using K-Means clustering (S421); Document clustering of detailed schedules included in the new construction plan information and existing construction plan information into fifth clusters using K-Means clustering (S422); A step of designating the average value of the 'order quantity by type of construction bolts included in the fourth cluster' as the 2-1 partial demand based on 'order receipt information for construction bolts of construction companies included in the fourth cluster' (S423); A step of designating the average value of the 'order quantity by type of construction bolts included in the 5th cluster' as the 2-2 partial demand based on 'order receipt information for construction bolts of construction companies included in the 5th cluster' (S424); and a step (S425) of designating an average value of the 2-1 partial demand quantity and the 2-2 partial demand quantity as the second partial demand quantity.

In the step (S421) of document clustering the construction plans included in the new construction plan information and the existing construction plan information into fourth clusters using K-Means clustering, as in a general document clustering algorithm, stopword removal, word extraction, Construction plans (documents) can be clustered through processes such as word vectorization (TFIDF, etc.).

In the step (S422) of document clustering the detailed schedules included in the new construction plan information and the existing construction plan information into fifth clusters using K-Means clustering, as in a general document clustering algorithm, stopword removal, word extraction, Detailed process tables (documents) can be clustered through processes such as word vectorization (TFIDF, etc.).

Here, the detailed process table may cluster documents based on words/characters extracted from the documents.

A step of designating the average value of the 'order quantity by type of construction bolts included in the fourth cluster' as the 2-1 partial demand based on 'order receipt information for construction bolts of construction companies included in the fourth cluster' (S423); In the second- 1 Partial demand can be calculated.

The average value of the 2-1 partial demand for each type of construction bolt calculated in this way and the 2-2 partial demand described later can be added to calculate the demand for construction bolts by applying a predetermined weight, for example, a weight of 0.3. there is.

A step of designating the average value of the 'order quantity by type of construction bolts included in the 5th cluster' as the 2-2 partial demand based on 'order receipt information for construction bolts of construction companies included in the 5th cluster' (S424); In the second- 2 Partial demand can be calculated.

In the step (S425) of designating the average value of the 2-1 partial demand and the 2-2 partial demand as the second partial demand, based on the 2-1 partial demand calculated based on the construction plan and the detailed schedule The average value of the calculated 2-2 partial demand is designated as the second partial demand.

The second partial demand for each type of construction bolt calculated in this way can be added to calculate the demand for construction bolts by applying a predetermined weight, for example, a weight of 0.3.

The step of calculating the third partial demand (S430) is: from the inventory change information, 'Month' as 'column', the type of construction bolt as 'row', and monthly inventory by type of construction bolt. Generating a third matrix with 'value' (S431); Based on the third matrix, clustering the types of construction bolts into sixth clusters using K-Means clustering for monthly inventory by type of construction bolts (S432); And, based on the monthly inventory by type of construction bolts included in the sixth cluster, designating the average value of 'monthly inventory by type of construction bolts included in the sixth cluster' as the third partial demand (S433); You can.

From the inventory change information, generating a third matrix with 'Month' as 'column', the type of construction bolt as 'row', and the monthly inventory for each type of construction bolt as 'value' ( In S431), a third matrix regarding monthly inventory of construction bolts for each month (January to December) is generated based on data accumulated over several years.

For example, in the third matrix, the data for 'A' construction bolts is that the 'March' inventory over the past 5 years is 10, the 'April' inventory over the past 5 years is 60, and the 'May' inventory over the past 5 years is 100. can be reflected.

Based on the third matrix, clustering the types of construction bolts into sixth clusters using K-Means clustering for the monthly inventory by type of construction bolts (S432); In the step (S432), monthly inventory by type of each construction bolt is Based on this, the types of construction bolts can be classified into the 6th cluster.

For example, 'A' construction bolts have a March inventory of 100 units, August inventory of 500 units, and December inventory of 50 units, and 'B' construction bolts have a March inventory of 90 units and August inventory of 520 units. , if the December inventory is 30, the inventory trends of 'A' construction bolts and 'B' construction bolts are similar, so they can be classified into the same 6th cluster.

Based on the monthly inventory by type of construction bolts included in the sixth cluster, the average value of 'monthly inventory by type of construction bolts included in the sixth cluster' is designated as the third partial demand (S433); 6 Based on the construction bolt order information for the construction projects included in the cluster, that is, the average value of the shipment quantity (order quantity) by type of existing construction bolts included in the existing construction plan information, the third part demand predicted to be ordered in the new construction can be calculated. You can.

The third partial demand for each type of construction bolt calculated in this way can be added to calculate the demand for construction bolts by applying a predetermined weight, for example, a weight of 0.4.

In addition, the step of calculating the first partial demand (S410) includes: selecting a reference target construction based on the existing construction plan information; Extracting the existing construction start month from the existing construction plan information corresponding to the reference target construction; extracting a new construction start month from the new construction plan information; Calculating a difference between the existing construction start month and the new construction start month; calculating a first weighting ratio by dividing the difference by 12; and recalculating the first partial demand by subtracting a value obtained by multiplying the first partial demand by the first weighting rate from the first partial demand.

In the step of selecting the reference target construction based on the existing construction plan information, the manager may select the existing construction plan information of the existing construction that is highly correlated with the new construction as the reference target construction.

In the step of extracting the existing construction start month from the existing construction plan information corresponding to the reference target construction, the construction start month of the reference target construction is extracted.

In the step of extracting the new construction start month from the new construction plan information, the construction start month of the new construction is extracted.

In the step of calculating the difference between the existing construction start month and the new construction start month, the difference value is calculated by comparing the existing construction start month and the new construction start month.

Here, the difference value is: If the start month of the existing construction is a larger value than the start month of the new construction, it can be calculated by specifying the start year of the existing construction one year earlier than the start year of the new construction.

For example, if the start month of existing construction is March and the start month of new construction is May 2023, the start month of existing construction is smaller than the start month of new construction (3 < 5), so the difference value is calculated as May - March = 2,

If the start month of existing construction is May and the start month of new construction is March 2023, the start month of existing construction is greater than the start month of new construction (5 > 3), so the difference value is May 2022 - It can be calculated as March 2023 = 10.

In the step of calculating a first weighting rate by dividing the difference value by 12, the first weighting rate is calculated by dividing the difference value by 12. Accordingly, the maximum value of the first weighting ratio may be 11/12 and the minimum value may be 0.

recalculating the first partial demand by subtracting a value obtained by multiplying the first partial demand by the first weighting rate from the first partial demand; correcting the first partial demand based on the first weighting rate; do.

In other words, changes in construction characteristics or demand for construction bolts according to the start month of construction can be reflected in the first partial demand based on the start month of existing construction.

The step of calculating the second partial demand (S420) includes: extracting an existing construction plan and an existing detailed schedule from existing construction plan information corresponding to the reference target construction; Extracting a new construction plan and a new detailed schedule from the new construction plan information; Calculating a first document similarity between the existing construction plan and the new construction plan; calculating a second document similarity between the existing detailed schedule and the new detailed schedule; Calculating a second weighting rate by adding 0.1 to the first document similarity, and correcting the calculated second weighting rate to 1 if it exceeds 1; Calculating a third weighting ratio by adding 0.1 to the second document similarity, and correcting the calculated third weighting ratio to 1 if it exceeds 1; recalculating the 2-1 partial demand by multiplying the 2-1 partial demand by the second weighting factor; and recalculating the 2-2 partial demand by multiplying the 2-2 partial demand by the third weighting factor.

In the step of extracting an existing construction plan and an existing detailed schedule from the existing construction plan information corresponding to the reference target construction, the construction plan and a detailed schedule of the reference target construction are extracted.

In the step of extracting a new construction plan and a new detailed schedule from the new construction plan information, the construction plan and a detailed schedule of the new construction are read.

In the step of calculating the first document similarity between the existing construction plan and the new construction plan, the first document similarity between the existing construction document and the new construction document is calculated.

In the step of calculating the second document similarity between the existing detailed schedule and the new detailed schedule, the second document similarity is calculated between the existing construction document and the new construction document.

Here, the first document similarity and the second document similarity can be calculated based on a general document similarity calculation algorithm.

Calculating a value obtained by adding 0.1 to the first document similarity as a second weighting rate, and correcting the calculated second weighting rate to 1 if it exceeds 1; In the step, correcting the first document similarity to a second weighting rate Calculate .

Calculating a value obtained by adding 0.1 to the second document similarity as a third weighting rate, and correcting the calculated third weighting rate to 1 if it exceeds 1; In the step, correcting the second document similarity to a third weighting rate Calculate .

In general, since there are bound to be differences between existing construction documents and new construction documents, 0.1 (10%) is added to the first document similarity and the second document similarity as above to calculate the second and third weighting rates. .

recalculating the 2-1 partial demand by multiplying the 2-1 partial demand by the second weighting factor; and recalculating the 2-2 partial demand by multiplying the 2-2 partial demand by the third weighting factor; The 2-1 partial demand quantity and the 2-2 partial demand quantity are corrected by reflecting the 2 weighting factor and the 3rd weighting factor.

In addition, the step of predicting the demand for construction bolts (S400) includes: calculating a root mean square value for the amount of inventory change by type of construction bolts from the monthly inventory of each type of construction bolts; For each type of construction bolt, calculating a first frequency at which the inventory change amount for each type of construction bolt exceeds the root mean square value; Calculating a change in inventory by type of construction bolts over the past year from the monthly inventory by type of construction bolts; Calculating a second frequency at which the change in inventory for each type of construction bolt over the past year exceeds the root mean square value; And when the second frequency is greater than or equal to the first frequency: in the demand quantity for construction bolts, a step of correcting the order quantity for each type of construction bolt with the root mean square value.

In the step of calculating the root mean square value for the inventory change amount by type of construction bolt from the monthly inventory by type of construction bolt, based on data accumulated over several years (monthly average value), monthly inventory change amount by type of construction bolt Calculate the root mean square value of .

Here, the increased and decreased values based on the average value of the monthly inventory are calculated as the inventory change amount, and the root mean square is calculated based on this.

For example, if the inventory of 'A' construction bolts from January to December is 176, 223, 487, 119, 224, 187, 487, 207, 448, 186, 462, and 420, the average value is 302 (decimal point) Based on (rounded to the 1st place), the inventory changes for 'A' construction bolts from January to December are -126, -79, 185, -183, -78, -115, 185, -95, 146, respectively. It can be calculated as -116, 160, and 118.

At this time, the root mean square value for the 'A' construction bolt based on the above '-126, -79, 185, -183, -78, -115, 185, -95, 146, -116, 160, 118' can be calculated. In the example data, the root mean square value can be calculated as 137.

In the step of calculating, for each type of construction bolt, a first frequency at which the inventory change amount for each type of construction bolt exceeds the root mean square value, the number of times the inventory change amount exceeds the root mean square value is counted.

In the '-126, -79, 185, -183, -78, -115, 185, -95, 146, -116, 160, 118' example, 4 times 185, 185, 146, 160 (exceeding 137) ) can be calculated as the first frequency.

Calculating the inventory change amount by type of construction bolt for the past year from the monthly inventory by type of the construction bolt; In the step, based on only the data for the most recent year among the data, the change amount of stock (monthly) by type of the construction bolt for the past year Calculate .

In the step of calculating a second frequency at which the inventory change amount by type of the construction bolt over the past year exceeds the root mean square value, the inventory change amount by type of construction bolt over the past year is calculated based on accumulated data. The second frequency, which is the number of times the root mean square value is exceeded, is calculated.

When the second frequency is greater than or equal to the first frequency: in the demand quantity of construction bolts, correcting the order quantity for each type of construction bolts with the root mean square value; in the second frequency calculated based on data of the most recent year Determine whether exceeds the first frequency calculated based on accumulated data.

Here, when the second frequency is higher than the first frequency, the frequency of excessive orders for construction bolts of the corresponding type is already high, so by limiting the order quantity to the root mean square value and correcting it, the occurrence of malicious inventory due to overordering can be prevented.

On the other hand, if the second frequency is less than the first frequency, the frequency with which the corresponding type of construction bolt is excessively ordered is not yet high, and the previously calculated order quantity is used as is to order the corresponding type of construction bolt.

The embodiments described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, and a field programmable gate (FPGA). It may be implemented using one or more general-purpose or special-purpose computers, such as an array, programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

Although the embodiments have been described with limited drawings as described above, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments, and equivalents of the claims also fall within the scope of the following claims.

S100: Step of receiving new construction plan information
S200: Step of receiving order information for construction bolts
S300: Step of recording inventory change information of construction bolts
S310: Step of measuring the weight of each type of construction bolt
S320: Step of recording the weight of each type of construction bolt by time
S330: Step of calculating inventory change over time by type of construction bolt
S400: Step to predict demand for construction bolts
S410: Calculating the first partial demand quantity
S411: Step of generating second matrix
S412: Step of standardizing the values of the second matrix by dividing them by the average value for each column
S413: Clustering construction projects into second clusters
S414: Step to designate the third cluster containing new construction
S415: Step of designating the average value of 'order quantity by type of construction bolt included in the third cluster' as the first partial demand quantity
S420: Step of calculating second partial demand
S421: Document clustering of construction plans into fourth clusters using K-Means clustering
S422: Step of document clustering of detailed schedules into fifth clusters using K-Means clustering
S423: Step of designating the average value of 'order quantity by type of construction bolt included in the 4th cluster' as the 2-1 partial demand quantity
S424: Step of designating the average value of 'order quantity by type of construction bolt included in the 5th cluster' as the 2-2 partial demand quantity
S425: Step of designating the average value of the 2-1 partial demand quantity and the 2-2 partial demand quantity as the second partial demand quantity
S430: Step of calculating third partial demand
S431: Step of generating third matrix
S432: Step of clustering types of construction bolts into 6th clusters
S433: Step of designating the average value of 'monthly inventory by type of construction bolts included in the 6th cluster' as the third part demand quantity
S440: Step of calculating the demand for construction bolts
S500: Step of ordering construction bolts
S600: Step of moving construction bolts inside the warehouse
S610: Generating the first matrix
S620: Step of clustering types of construction bolts into first clusters
S630: Step of distributing and designating storage locations for each type of construction bolt to different shelves
S640: Step of controlling a certain AGV to transfer trays

Claims (3)

In the method of predicting demand and inventory management of construction bolts based on analysis of construction plan information performed by the device,
Step of receiving new construction plan information;
Step of receiving order information for construction bolts;
Recording inventory change information of the construction bolts;
Predicting the demand for the construction bolts based on new construction plan information, existing construction plan information, and inventory change information; and
A step of ordering construction bolts based on the predicted demand for construction bolts,
The steps to predict the demand for construction bolts are:
calculating a first partial demand based on the new construction plan information and the existing construction plan information;
calculating a second partial demand based on the new construction plan information and the existing construction plan information;
calculating a third partial demand based on the inventory change information; and
Comprising: calculating the demand for construction bolts by reflecting a predetermined weight for each of the first to third partial demands and adding them up;
The step of calculating the first partial demand is:
Creating a second matrix with the construction price included in the new construction plan information and the existing construction plan information, the correlation by construction type, the construction start month, and the construction period as 'columns', and the unique number of the construction as 'rows'. step;
standardizing the values of the second matrix by dividing them by the average value for each column;
Based on the standardized second matrix, clustering the construction projects into second clusters using K-Means clustering for the new construction plan information and the existing construction plan information;
Among the second clusters, designating a third cluster containing new construction; and
A step of designating the average value of the 'order quantity by type of construction bolts included in the third cluster' as the first partial demand based on 'order information for construction bolts of construction companies included in the third cluster',
The step of calculating the second partial demand is:
Document clustering the construction plans included in the new construction plan information and the existing construction plan information into fourth clusters using K-Means clustering;
Document clustering the detailed schedules included in the new construction plan information and the existing construction plan information into fifth clusters using K-Means clustering;
Designating the average value of the 'order quantity by type of construction bolts included in the fourth cluster' as the 2-1 partial demand based on 'order receipt information for construction bolts of construction companies included in the fourth cluster';
Designating the average value of the 'order quantity by type of construction bolts included in the 5th cluster' as the 2-2 partial demand based on 'order receipt information for construction bolts of construction companies included in the 5th cluster'; and
A step of designating the average value of the 2-1 partial demand quantity and the 2-2 partial demand quantity as the second partial demand quantity,
The steps for calculating the third partial demand are:
From the inventory change information, generating a third matrix with 'Month' as 'column', the type of construction bolt as 'row', and the monthly inventory for each type of construction bolt as 'value';
Based on the third matrix, clustering the types of construction bolts into sixth clusters using K-Means clustering for monthly inventory by type of construction bolts; and
Based on the monthly inventory by type of construction bolts included in the sixth cluster, designating the average value of 'monthly inventory by type of construction bolts included in the sixth cluster' as the third partial demand; including,
Demand forecasting and inventory management method for construction bolts based on analysis of construction plan information In claim 1,
The construction bolts are: stored inside a designated warehouse,
A plurality of shelves are arranged inside the warehouse, a plurality of trays are placed on the shelves, and the construction bolts are accommodated in different trays for each type,
It further includes the step of moving the construction bolts inside the warehouse,
The steps for recording inventory change information for construction bolts are:
Measuring the weight of each type of construction bolt from a weight sensor installed on a tray accommodating each type of construction bolt;
Recording the weight of each type of construction bolt by time; and
Comprising: calculating a change in inventory by time by type of the construction bolt, based on the weight by time by type of the construction bolt,
The steps of moving the construction bolts inside the warehouse where the construction bolts are stored are:
Creating a first matrix with the time zone as 'column', the type of construction bolt as 'row', and the inventory change by time zone as 'value';
Based on the first matrix, clustering the types of construction bolts into first clusters using K-Means clustering for the amount of change in inventory by time period for each type of construction bolts;
Distributing and designating storage locations for each type of construction bolts included in each first cluster to different shelves; and
Including, controlling a predetermined AGV to transfer the tray according to the storage location for each type of construction bolt.
Demand forecasting and inventory management method for construction bolts based on analysis of construction plan information delete