Inventory placement plan determination device

Abstract

To accurately predict timing at which a demand for a maintenance component is generated and enable stock management according to the timing.SOLUTION: A stock arrangement plan determination apparatus comprises: a storage device which includes a storage unit which stores various information; and a control device which includes a calculation unit which executes various calculations about arrangement of stock of a component on the basis of the information stored in the storage unit. The storage unit stores at least demand information about the demand for the component and retrieval history information about a history of retrieval about the component. The calculation unit comprises: a demand generation timing prediction part which predicts timing at which the demand for the component is generated in the future on the basis of at least the demand information and retrieval history information; and a stock arrangement plan generation part which generates a stock arrangement plan for stock of the component in a physical distribution base on the basis of the prediction result by the demand generation timing prediction part.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to an inventory allocation plan determination device.

Background Art

[0002] In the operation of construction machinery, since the arrival of the life of maintenance parts, failures, and wear occur regularly or irregularly, construction machinery manufacturers and dealerships are required to supply such maintenance parts to customers quickly.

[0003] Among maintenance parts, there are parts that reach the end of their life in a relatively short cycle and are frequently replaced, while there are also many parts for which demand occurs only about once every few years. Such parts for which demand occurs only once every few years are generally expensive. If such expensive maintenance parts with low demand are always held in a warehouse (regional depot) near the end-user (customer)'s base, those maintenance parts will occupy the warehouse for several years, resulting in the occurrence of storage management operations, a decrease in inventory turnover rate, and an increase in inventory assets. Therefore, management is required to not hold such low-demand and high-cost maintenance parts as inventory as much as possible.

[0004] On the other hand, from the perspective of improving serviceability, it is desirable to suppress the possibility of the end-user's business being stopped due to the stoppage of construction machinery operation as much as possible. For this purpose, it is required to hold even low-demand and high-cost maintenance parts in a logistics base (regional depot) as close as possible to the end-user's base, realize immediate shipment upon receipt of an order, and shorten the delivery lead time.

[0005] In order to improve serviceability while reducing the inventory of maintenance parts, a technique for predicting the demand for maintenance parts and using it for inventory management is known, for example, from Patent Document 1. The technique of Patent Document 1 predicts future demand using information such as item characteristics, past demand records, and past operation records of machinery, and manages inventory according to the prediction results.

Prior Art Documents

Patent Documents

[0006] [Patent Document 1] Japanese Patent Publication No. 2006-85645 [Overview of the project] [Problems that the invention aims to solve]

[0007] However, conventional technologies often fail to show a correlation between past demand and past machine operating performance, resulting in insufficient forecasting accuracy.

[0008] This disclosure is made in light of the circumstances described above, and provides an inventory allocation plan determination device that accurately predicts the timing of demand for maintenance parts and enables inventory management in line with that timing. [Means for solving the problem]

[0009] To solve the above problems, the inventory placement plan determination device according to this disclosure comprises a control device and a storage device, and is an inventory placement plan determination device for determining the placement of parts inventory at a logistics base, wherein the storage device comprises a storage unit for storing various information, and the control device comprises a calculation unit for performing various calculations related to the placement of the parts inventory based on the information stored in the storage unit. The storage unit stores at least demand information regarding the demand for the parts and search history information regarding the search history of the parts. The calculation unit comprises a demand occurrence timing prediction unit that predicts the timing at which demand for the parts will occur in the future based at least on the demand information and the search history information, and an inventory placement plan generation unit that generates an inventory placement plan for the parts inventory at the logistics base based on the prediction result of the demand occurrence timing prediction unit. [Effects of the Invention]

[0010] According to this disclosure, it is possible to provide an inventory allocation plan determination device that can accurately predict when demand for maintenance parts will arise and enable inventory management in line with that timing. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram illustrating the inventory placement plan determination device 100 according to the first embodiment. [Figure 2] This is a data structure diagram showing an example of the data structure of the item master information storage unit 111. [Figure 3] This is a data structure diagram showing an example of the data structure of the demand information storage unit 112. [Figure 4] This is a data structure diagram showing an example of the data structure of the inventory information storage unit 113. [Figure 5] This is a data structure diagram showing an example of the data structure of the inventory placement plan information storage unit 114. [Figure 6] This is a data structure diagram showing an example of the data structure of the search history information storage unit 115. [Figure 7] This is a data structure diagram showing an example of the data structure of the machine operation information storage unit 116. [Figure 8] This is a data structure diagram showing an example of the data structure of the prediction result information storage unit 117. [Figure 9] This is a data structure diagram showing an example of the data structure of the evaluation index calculation result information storage unit 118. [Figure 10] This flowchart illustrates an example of a processing procedure performed by the inventory placement plan determination device 100 according to the first embodiment. [Figure 11] An example of output screen 1400 is shown. [Figure 12] This is a flowchart illustrating the operation of the inventory placement plan determination device 100 according to the second embodiment. [Figure 13] This is a schematic diagram illustrating the basic principle of the inventory placement plan determination device 100. [Figure 14] This is a schematic diagram illustrating the basic principle of the inventory placement plan determination device 100. [Modes for carrying out the invention]

[0012] Hereinafter, this embodiment will be described with reference to the attached drawings. In the attached drawings, functionally identical elements may be denoted by the same reference numerals. Note that the attached drawings illustrate embodiments and implementation examples in accordance with the principles of the present disclosure, but these are for the purpose of understanding the present disclosure and are not used to limit the interpretation of the present disclosure in any way. The description in this specification is merely a typical example and does not limit the claims or application examples of the present disclosure in any sense.

[0013] In this embodiment, the description is provided in sufficient detail for those skilled in the art to implement the present disclosure. However, other implementations and forms are possible, and it is necessary to understand that changes in configuration and structure and replacement of various elements can be made without departing from the scope and spirit of the technical idea of the present disclosure. Therefore, the following description should not be construed as being limited thereto.

[0014] [First Embodiment] Hereinafter, the inventory placement plan determination device 100 according to the first embodiment will be described with reference to FIG. 1 and the like. Before entering the description of the specific device configuration, the premise of this inventory placement plan determination device 100 will be described with reference to FIGS. 13 and 14.

[0015] As shown in FIG. 13, the maintenance parts of construction machinery are manufactured by a supplier (e.g., a construction machinery manufacturer) and supplied to an end user (such as a construction company) through a central logistics base (CPD), regional logistics bases (regional depots), dealerships, etc. The regional depot is located near the end user's activity base or near the dealership and supplies maintenance parts to the end user promptly. The CPD coordinates multiple regional depots, receives the supply of maintenance parts from the supplier, and appropriately supplies and stores the maintenance parts in multiple regional depots.

[0016] Many of these maintenance parts are low-demand items that are only needed once every few years, and these items tend to be expensive. If such low-demand, high-priced maintenance parts are kept in stock at frontline regional depots, they will occupy warehouse space for several years, resulting in storage management work, a decrease in inventory turnover, and an increase in inventory assets. For this reason, it is desirable to avoid keeping such low-demand, high-priced maintenance parts in stock as much as possible.

[0017] However, from the perspective of improving serviceability, it is necessary to have comprehensive maintenance contract terms such as FMC and to minimize the disruption of end-user operations due to machine downtime. To achieve this, it is desirable to stock such low-demand, high-cost maintenance parts at frontline regional depots, enable same-day shipment upon order, and shorten delivery times.

[0018] Therefore, for such low-demand, high-priced maintenance parts, it is preferable to appropriately switch between three states, as shown in Figure 14, for example: a state where inventory is increased at frontline regional depots (normal frontline), a state where inventory is decreased at frontline regional depots and the proportion of inventory at rear-end CPDs is increased (frontline reduction), and a state where inventory is not kept at frontline regional depots but is held only at rear-end CPDs (centralized). By appropriately adjusting these three states based on the forecast of demand for low-demand, high-priced maintenance parts, it is possible to improve serviceability while suppressing inventory management costs. However, it has been difficult to accurately predict the occurrence of demand for maintenance parts and to implement such adjustments while avoiding parts shortages and excess inventory. The inventory allocation plan determination device 100 of the first embodiment accurately performs demand forecasting by the method detailed below, thereby generating inventory allocation plans for CPDs and regional depots.

[0019] Referring to Figure 1, the inventory placement plan determination device 100 according to the first embodiment will be described. The inventory placement plan determination device 100 of the first embodiment is connected via a network NW to a user terminal 200 used by the user and to a database 300 where various data are stored.

[0020] The inventory placement plan determination device 100 is an information processing device such as a personal computer or a server computer, and for example comprises a storage unit 110, an arithmetic unit 120, an input unit 130, and an output unit 140. Specifically, the computer comprising the inventory placement plan determination device 100 may include, for example, a CPU (Central Processing Unit) 101, a GPU (Graphics Processing Unit) 102, a ROM 103, RAM 104, a hard disk drive (HDD) 105, a display control unit 106, an input / output control unit 107, and a communication control unit 108, in addition to a display, keyboard, mouse, etc. The CPU 101 is a central control unit that manages the overall operation of the inventory placement plan determination device. The GPU 102 is, for example, a control unit for performing image processing of image data.

[0021] ROM 103 is a storage device that stores various data necessary for executing various programs. RAM 104 is a storage device that temporarily stores the calculation results of the same programs. Programs stored in ROM 103 include, for example, training programs for executing the training process of a classifier. The hard disk drive 105 is also a storage device and can store trained models constructed by machine learning processing. ROM 103, RAM 104, and hard disk drive 105 together constitute the storage unit 110. Various programs may also be stored in a portable storage medium (e.g., CD-ROM) 109 that is readable by a computer.

[0022] The display control unit 106 is a control unit and control device responsible for displaying the execution screen of the aforementioned program, etc., on a display (not shown). The input / output control unit 107 is a control unit and control device that controls the input of data and instructions from various input devices, and the output of various data output from the CPU 101 or GPU 102. The communication control unit 108 is a control unit and control device that manages data communication with an external computer.

[0023] The user terminal 200 is an information processing device such as a personal computer or tablet. The user issues instructions to the inventory placement plan determination device 100 to execute processing via the user terminal 200. The user terminal 200 also has the function of displaying information output by the inventory placement plan determination device 100 to the user via a display, printer, etc. (not shown).

[0024] The database 300 is, for example, a database or storage device that stores data such as an ERP (Enterprise Resource Planning) system or similar data. The network NW connects the user terminal 200, the database 300, and the inventory placement plan determination device 100 in a communication manner. The network NW is, for example, a communication network that uses public lines such as a LAN (Local Area Network), WAN (Wide Area Network), VPN (Virtual Private Network), or the Internet, either partially or entirely.

[0025] The memory unit 110 is a memory unit that stores various data used to determine the inventory placement plan for maintenance parts, and includes, as an example, the following storage units. Details of each storage unit will be described later. • Item master information storage unit 111, which stores information regarding the characteristics of maintenance parts. • Demand information storage unit 112 stores information regarding the demand for maintenance parts registered in the item master information storage unit 111. • Inventory information storage unit 113, which stores information on the historical inventory levels of specific maintenance parts at each logistics base. • Search history information storage unit 115, which stores the search history of end users (customers) on mail-order sites regarding maintenance parts. • Machine operation information storage unit 116, which stores information on the cumulative operating time of construction machinery. • Prediction result information storage unit 117 that stores the prediction results from the demand generation timing prediction unit 121. • Evaluation index calculation result information storage unit 118 for storing the calculation results by the evaluation index calculation unit 123.

[0026] The calculation unit 120 is a calculation unit for calculating the timing of demand for maintenance parts, generating inventory placement plans based on that timing, and performing calculations related to evaluation indicators. For example, it includes a demand timing forecasting unit 121, an inventory placement plan generation unit 122, and an evaluation indicator calculation unit 123.

[0027] The demand timing forecasting unit 121 forecasts the timing of future demand for each maintenance part based on information on past demand, information on the search history of maintenance parts, etc., and stores the forecast result in the forecast result information storage unit 117. The inventory placement plan generation unit 122 generates an inventory placement plan for the maintenance parts related to the forecast based on the forecast result (timing of future demand for each maintenance part) stored in the forecast result information storage unit 117, and stores the generated result in the inventory placement plan information storage unit 114. The evaluation index calculation unit 123 reads various information from the demand information storage unit 112, inventory information storage unit 113, inventory placement plan information storage unit 114, and forecast result information storage unit 117, calculates the service rate, inventory value, and expected number of orders for each location and each combination of maintenance parts as evaluation indexes, and stores the calculation result in the evaluation index calculation result information storage unit 118.

[0028] Figure 2 shows an example of the data structure of the item master information storage unit 111. The item master information storage unit 111 stores information about the characteristics of maintenance parts, and may consist of data items such as part number, characteristic classification, weight, size, compatible model, and quantity used. The part number is information about the part number (identification number) used to identify the maintenance part. The characteristic classification is classification information that indicates the characteristics of the maintenance part (for example, hydraulic system parts, electrical system parts, consumable parts, etc.). The weight indicates the weight of the maintenance part. The size indicates the size of the maintenance part (an index indicating size, the volume occupied by the part, etc.). The compatible model is information about the model of construction machinery in which the maintenance part is used. The quantity used indicates the quantity of the maintenance part to be used in the compatible model. The information in the first row of Figure 2 shows that the hydraulic system maintenance part with part number 001 has a weight of 5 kg, a size of 0.01, is compatible with model A, and has a quantity used of 1.

[0029] Figure 3 shows an example of the data structure of the demand information storage unit 112. The demand information storage unit 112 stores information regarding the demand for maintenance parts registered in the item master information storage unit 111, and consists of data items such as the date the demand occurred, the part number, the quantity, and the order location. For example, the information in the first row of Figure 3 indicates that on January 1, 2019, an order for one maintenance part with part number 001 was received from location 1 (demand occurred).

[0030] Figure 4 shows an example of the data structure of the inventory information storage unit 113. The inventory information storage unit 113 stores information on the historical inventory levels of specific maintenance parts at each logistics base. For example, it may include the base name, part number, date, inventory quantity, and inventory unit price as data items. The base name indicates the name of the logistics base (regional depot) that holds the inventory. The part number indicates the identification number of the maintenance part that was held in inventory. The date indicates the date the inventory was confirmed. The inventory quantity indicates the number of maintenance parts in stock at the logistics base (regional depot). The inventory unit price indicates the inventory unit price of the maintenance part at the logistics base.

[0031] Figure 5 shows an example of the data structure of the inventory placement plan information storage unit 114. The inventory placement plan information storage unit 114 stores information regarding inventory placement rules (placement plans) at each logistics base, and may include the base name, part number, forecast score, and placement rule as data items. The base name indicates the name of the regional depot to which the inventory placement plan is applied. The part number indicates the identification number of the maintenance part to which the inventory placement plan is applied. The forecast score is a score used to determine the inventory placement plan, and is calculated based on past demand performance as well as the search history for the maintenance part in question. The placement rule indicates the rules for inventory placement, and can be classified, for example, into "Frontline Normal," "Frontline Reduced," "Centralized," etc.

[0032] Figure 6 shows an example of the data structure of the search history information storage unit 115. The search history information storage unit 115 stores the search history of the mail-order website used by end users (customers) when purchasing maintenance parts registered in the item master information storage unit 111 from construction machinery manufacturers. The mail-order website may be operated directly by the construction machinery manufacturer, or it may be operated by an intermediary, sales agent, transportation company, etc., that partners with the construction machinery manufacturer. End users purchasing maintenance parts search for the maintenance parts they wish to purchase on the aforementioned mail-order website, investigate the delivery date and price, enter the quantity to be purchased, and place an order. Therefore, the timing of the occurrence of demand for maintenance parts can be estimated according to the search history on the mail-order website.

[0033] The search history information storage unit 115 may include, as an example, a location name, a part number, and a date and time as data items. The location name indicates the name of the regional depot that receives orders from customers and ships the ordered maintenance parts. The part number indicates the part number of the maintenance part searched on the aforementioned e-commerce site. The date and time indicates the time information when the search for the maintenance part was performed on the aforementioned e-commerce site.

[0034] Figure 7 shows an example of the data structure of the machine operation information storage unit 116. The machine operation information storage unit 116 stores information on the type of construction machine and the cumulative operating time for each machine. Specifically, it may include data items such as the type name, machine number, activity base name, date, and cumulative operating time. The type name indicates the name of the type of construction machine. The machine number indicates the identification number assigned to distinguish multiple construction machines of the same type. The activity base name indicates the activity area (construction site, etc.) where each construction machine is operating. In the example in Figure 7, the activity base name is indicated by the name of the regional depot that supplies maintenance parts to the activity area. The date indicates the date on which the cumulative operating time of the construction machine reached the value indicated by the cumulative operating time. The cumulative operating time is a numerical value indicating the cumulative operating time of the construction machine.

[0035] Figure 8 shows an example of the data structure of the forecast result information storage unit 117. The forecast result information storage unit 117 stores information indicating the results of demand occurrence timing forecasts calculated in the past by the demand occurrence timing forecasting unit 121. Specifically, as an example, it may include the location name, part number, forecast date, and two-year demand occurrence probability as data items. The location name indicates the regional depot that was the target of the demand forecast. The part number indicates the maintenance part that was the target of the demand forecast. The forecast date indicates the date on which the demand occurrence timing forecasting unit 121 performed the demand forecast. The two-year demand occurrence probability indicates, as predicted by the demand occurrence timing forecasting unit 121, whether or not demand will occur within two years from the forecast date, using the numerical value "1" or "0". "0" means that it is predicted that no demand will occur within two years, and "1" means that it is predicted that no demand will occur within two years. For example, the information in the first row of Figure 8 indicates that the probability of demand for "Maintenance Part 001" at "Location A" within two years from the predicted date (January 1, 2020) is "1," meaning that demand is predicted to occur.

[0036] Figure 9 shows an example of the data structure of the evaluation index calculation result information storage unit 118. The evaluation index calculation result information storage unit 118 stores information on the calculation results of the evaluation index calculated by the evaluation index calculation unit 123. Specifically, as an example, it may include the location name, part number, date, service rate, inventory value, and estimated number of orders as data items. The location name indicates the name of the regional depot that was the subject of the evaluation index calculation. The part number indicates the part number of the maintenance part that was the subject of the evaluation index calculation. The date indicates the date on which the evaluation index was calculated. The service rate indicates the estimated result of the service rate of the relevant maintenance part at the relevant location (regional depot) (the degree to which delivery is possible within a predetermined period after an order is placed). The inventory value indicates the estimated result of the inventory value of the relevant maintenance part at the relevant location (regional depot). The estimated number of orders indicates the prediction of how many times orders for the relevant maintenance part will occur at the relevant location (regional depot).

[0037] Next, with reference to the flowchart in Figure 10, an example of the processing procedure performed by the inventory placement plan determination device 100 in the first embodiment will be described. The processing procedure described below assumes that a predetermined number of item master information, demand information, inventory information, search history information, and machine operation information are recorded in the database 300, etc., and is started, for example, in response to a start command from the user to the user terminal 200.

[0038] First, in step S1, the input unit 130 acquires item master information, demand information, inventory information, search history information, and machine operation information from the database 300 via the network NW and stores (transfers) them to the storage unit 110.

[0039] Next, in step S2, the demand timing prediction unit 121 reads item master information, demand information, search history information, and machine operation information from the storage unit 110 and predicts the timing of demand for the maintenance part. Then, in step S3, the inventory placement plan generation unit 122 generates an inventory placement plan according to this prediction result.

[0040] The inventory placement plan generation unit 122 reads the forecast result information from the storage unit 110 and calculates a forecast score for each combination of location and part number, which is the sum of the two-year demand occurrence probabilities from the most recent six forecast results. If the forecast score is 3 or less, "Centralized Aggregation" is selected as the inventory placement plan for the location and part number combination. If the forecast score is 4 or more, "Frontline Normal" is selected as the inventory placement plan for the location and part number combination (the threshold for the forecast score can also be changed to select "Frontline Reduction" when the forecast score is an intermediate value). The combination of the forecast score and inventory placement plan calculated and generated for each location and part number combination is then stored in the inventory placement plan information storage unit 114. By using the results of multiple past forecasts to determine the inventory placement plan, as described above with the forecast score, it is possible to avoid situations where, for example, if monthly forecasts are made, the forecast results are reversed each month and the inventory placement plan is changed every month.

[0041] Next, in step S4, the evaluation indicator calculation unit 123 reads from the storage unit 110, the demand information storage unit 112, the inventory information storage unit 113, and the inventory placement plan information storage unit 114. Based on the inventory placement plan information storage unit 114, the evaluation indicator calculation unit 123 refers to the placement plan value for each combination of location and part number, and calculates the values ​​of two evaluation indicators, inventory value and service rate, in the following procedure.

[0042] If the proposed location is "centralized," it is predicted that no demand will be generated for the relevant location and part number combination over the next two years, so the inventory quantity of the relevant item at that location will not change. Therefore, the evaluation index calculation unit 123 obtains the latest inventory quantity and inventory unit price for the relevant location and part number combination from the inventory information storage unit 113, and calculates the inventory value by multiplying these values. Furthermore, regarding the service rate, since it is predicted that no demand will be generated for the next two years, the service rate will be set to 0% or excluded from the calculation. In addition, the expected number of orders will be set to 0.

[0043] If the deployment plan value is "Frontline Normal," it is predicted that demand may occur for the relevant location and part number combination within the next two years. Therefore, the evaluation index calculation unit 123 uses the past demand data for the relevant location and part number combination recorded in the demand information storage unit 112 to calculate the required inventory quantity based on general inventory theory, and places the corresponding number of maintenance parts as inventory at the regional depot. The evaluation index calculation unit 123 then obtains the latest inventory unit price for the relevant location and part number combination from the inventory information storage unit 113, and calculates the inventory value by multiplying this inventory unit price by the aforementioned required inventory quantity. As for the service rate, if the calculated inventory quantity is immediately available for deployment, it is set to 100%. Furthermore, using the past demand data for the relevant location and part number combination recorded in the demand information storage unit 112, the value of the number of demands for the most recent year can be used as the value of the expected number of orders. The evaluation index calculated in this manner (service rate, inventory value, expected number of orders) is output to a display or the like via the output unit 140 (step S5).

[0044] The prediction of demand timing in step S2 will be explained in more detail. The demand timing prediction unit 121 can sort the past demand data for maintenance parts recorded in the demand information storage unit 112 into a time series for each item and calculate the average value of the interval between demand occurrences. For example, if the interval between demand occurrences for a certain maintenance part is two years, and the most recent demand occurred one year ago, the demand timing prediction unit 121 can estimate that the next demand for that maintenance part will occur one year from now. In other words, it predicts that demand will occur at least once within the next two years. The result of this prediction can be stored in the demand timing prediction unit 121 as a prediction result based on demand data.

[0045] The demand timing prediction unit 121 can also link past demand data with search history information and predict the timing of demand occurrence according to the relationship between the two. There is a certain causal relationship between the past demand data for maintenance parts listed in the demand information and the past search history for maintenance parts on the mail-order site listed in the search history information. End users (customers) who purchase maintenance parts search for the maintenance parts they wish to purchase on the aforementioned mail-order site, investigate delivery dates and prices, enter the quantity to be purchased, and place an order. Therefore, when demand for maintenance parts occurs, there is a fairly high probability that a search has been conducted on the aforementioned mail-order site immediately beforehand, and that search history information exists. This kind of linking makes it possible to predict the timing of demand occurrence for maintenance parts with higher accuracy.

[0046] Therefore, in this embodiment, past search history stored in the search history information storage unit 115 and past demand records stored in the demand information storage unit 112 are associated (linked) by part number, the number of searches between the day a demand occurs and the day the next demand occurs is identified, and the average value of the time interval between those multiple searches is calculated. For example, suppose (1) the number of searches between the time a demand occurred and the present is 5, (2) the average value of the interval between those 5 searches is 30 days, and (3) 2 searches have been performed between the most recent demand and the present. In this case, it can be estimated that the period until the next demand occurs is within 90 days from the most recent demand. Therefore, the 2-year demand occurrence probability (Figure 8) is set to "1".

[0047] The demand timing prediction unit 121 can also link past search history with construction machinery operation information and predict the timing of demand occurrence according to the relationship between the two. A certain causal relationship can be recognized between past search history stored in the search history information storage unit 115 and construction machinery operation information stored in the machine operation information storage unit 116, similar to the relationship between search history and demand information. This is because as the frequency of use (operating time) of construction machinery by end users increases, they will consider that maintenance parts for the construction machinery need to be replaced, and will have more opportunities to search for such maintenance parts on the aforementioned mail-order site in order to purchase them or check their inventory. Therefore, in this embodiment, as an optional choice, in addition to (or instead of) predicting the timing of demand occurrence for maintenance parts based on the association between search history and demand information as described above, it is also possible to perform prediction of the timing of demand occurrence for maintenance parts based on the association between search history and construction machinery operation information.

[0048] Specifically, the search history information for maintenance parts whose demand timing is to be predicted is identified in the search history information storage unit 115, and the item master information storage unit 111 is referenced using the part number data of the identified search history information.

[0049] Then, the system searches for information corresponding to the part number in the item master information storage unit 111 and links the corresponding model information in the searched information with the original search history information. Using the linked corresponding model information and the activity base name information in the search history information storage unit 115, the system refers to the machine operation information storage unit 116 to identify the cumulative operating time for that corresponding model, and links the identified cumulative operating time with the original search history information. In other words, the search history information for a certain maintenance part is linked with the operation information of the construction machine in which that maintenance part is used.

[0050] Next, the cumulative operating time for each machine is sorted chronologically, and the average cumulative operating time between the most recent search and the previous search is calculated according to the linked search history. This allows us to calculate the number of days between the most recent search and the previous search, and the cumulative operating time during those days. As a result, we can estimate how many hours of cumulative operating time must be reached since the last search before the next search will be executed. For example, if the cumulative operating time between the most recent search and the previous search is 1000 hours, and the current cumulative operating time since the most recent search is 100 hours, then it can be predicted that the next new search will occur when the cumulative operating time has increased by another 900 hours.

[0051] Now, let's assume that the number of days since the most recent search is 300 days, when the cumulative operating time since the most recent search increases by another 900 hours. As in the example above, if five searches occur between demands, and two searches have been performed between the most recent demand and the present, then three more searches are needed before the next demand. In this case, taking into account the specific result of the cumulative operating time mentioned above, we can predict that the next demand will occur 3 x 300 days = 900 days later (approximately 2 years and 5 months later). In other words, we can predict that no demand will occur in the next two years, so the probability of demand occurring in two years can be set to "0". The probability of demand occurring in two years for each combination of location and item is calculated using the above procedure, and the current date and the calculation results are stored in the prediction result information storage unit 117.

[0052] As described above, the demand timing prediction unit 121 of this embodiment can more accurately predict the timing of demand by executing the demand timing calculation based on the actual demand for the maintenance part to be predicted and the search history of that maintenance part, thereby enabling the generation of an appropriate inventory allocation plan. Preferably, by also taking into account information on the operation information of the construction machinery in which the maintenance part to be predicted is used, and linking this with the search history information, the timing of demand for the maintenance part can be predicted even more accurately.

[0053] Furthermore, it is possible to independently perform two predictions: one based on search history and another based on past demand information. If both predictions indicate demand within two years, the probability of demand occurring within two years is set to "1," and otherwise, it is set to "0." Similarly, it is possible to independently perform three predictions: one based on search history, one based on past demand information, and one based on construction machinery operation information. If two of the three predictions indicate demand, the probability of demand occurring within two years is set to "1," and otherwise, it is set to "0." Alternatively, the final prediction can be obtained by weighting the three prediction results (e.g., 0.2x, 0.6x, etc.) and summing them, then rounding the result.

[0054] Figure 11 shows an example of the output screen 1400. The output screen 1400 consists, as an example, of an item-specific results display section 1401 and a site-specific results display section 1402. The item-specific results display section 1401 may include the characteristic classification, model, predicted score, inventory placement plan, estimated number of orders, service rate, and inventory value for each combination of site and part number. Here, the characteristic classification is the characteristic classification of the relevant part number as recorded in the item master information storage unit 111. The model is the corresponding model for the relevant part number as recorded in the item master information storage unit 111. The predicted score is the predicted score recorded in the inventory placement plan information storage unit 114 for the relevant site and part number combination. The inventory placement plan is the inventory placement plan recorded in the inventory placement plan information storage unit 114 for the relevant site and part number combination. The estimated number of orders is the estimated number of orders recorded in the evaluation index calculation result information storage unit 118 for the relevant site and part number combination. The service rate is the service rate recorded in the evaluation index calculation result information storage unit 118 for the relevant location and part number combination. The inventory value is the inventory value recorded in the evaluation index calculation result information storage unit 118 for the relevant location and part number combination.

[0055] The results display section 1402 for each location displays the aggregated results of the expected number of orders, service rate, and inventory value for each location. The results display section 1402 consists of the location, the expected number of orders, the service rate, and the inventory value. The location indicates the name of the location. The expected number of orders is the sum of the expected number of orders recorded in the evaluation index calculation result information storage unit 118 for the relevant location. The service rate is the result of multiplying the service rate for each item and the expected number of orders for each item at the relevant location, as recorded in the evaluation index calculation result information storage unit 118, and then summing these service rates to get the total number of service orders for each location. This total number of service orders is then divided by the aforementioned sum of expected number of orders. To give a specific example, the evaluation index calculation result information storage unit 118 in Figure 8 shows two pieces of information for location A, with part numbers 001 and 002, respectively. Here, part number 001 at site A has a service rate of 100% and an expected number of orders of 1, so the number of services is 100% × 1 = 1. Also, part number 002 at site A has a service rate of 0%, so the number of services is 0% × 0 = 0. In summary, the total number of services at site A, which is the sum of the number of services, is 0 + 1 = 1. On the other hand, the sum of the expected number of orders at site A is 1 + 0 = 1. Therefore, the service rate at site A is 1 ÷ 1 = 1, or 100%.

[0056] As explained above, the inventory placement plan determination device of the first embodiment predicts the timing of demand occurrence by taking into account not only actual demand but also the history of searches for maintenance parts. Therefore, it is possible to accurately predict the timing of demand occurrence and manage inventory in accordance with that timing.

[0057] [Second Embodiment] Next, with reference to Figure 12, the inventory placement plan determination device 100 according to the second embodiment will be described. The overall configuration of the device in this second embodiment may be the same as that of the first embodiment (Figure 1), and the data structure of the storage unit 110 may also be the same (Figures 2 to 9). Furthermore, the method for predicting the timing of demand occurrence and the generation of inventory placement plans may also be the same as in the first embodiment. This second embodiment is configured to exclude certain maintenance parts from the prediction of the timing of demand occurrence and the generation (switching) of inventory placement plans according to various characteristics of the maintenance parts (frequency of demand occurrence, importance, storage costs, etc.).

[0058] The operation of the second embodiment will be explained with reference to the flowchart in Figure 12. First, the calculation unit 120 identifies the part number of a maintenance part stored in the item master information storage unit 111 and refers to its demand information (step S11).

[0059] Then, the demand information is analyzed, the frequency of the demand is determined, and it is determined whether or not that frequency is greater than a predetermined threshold (step S12). If yes, the maintenance part is excluded from the inventory allocation plan change for this case, and a specific inventory allocation plan (e.g., frontline normal) is applied permanently (step S15). If no, proceed to step S13.

[0060] In step S13, the importance of the maintenance part is determined, and it is determined whether the importance index is above a predetermined threshold. If yes, the maintenance part is excluded from the inventory placement plan change for this case, and for example, the normal frontline inventory placement plan is fixed and applied (step S15). That is, the demand timing forecasting unit 121 does not forecast the demand timing for the maintenance part, and the generation and updating of the inventory placement plan by the inventory placement plan generation unit 122 is also stopped. If no, the process proceeds to step S14.

[0061] In step S14, the storage cost of the maintenance part is determined, and it is determined whether the storage cost is above a predetermined threshold. If the answer is No, a specific inventory placement plan (e.g., frontline normal) is fixedly applied to that maintenance part (step S15). If the answer is Yes, the process proceeds to step S16, where the normal inventory placement plan generation process (switching between frontline normal, frontline reduced, and centralized) is applied, similar to the first embodiment.

[0062] Thus, in the second embodiment, the application of the inventory allocation plan generation process of this disclosure is determined according to the characteristics of the maintenance parts. Therefore, according to this second embodiment, in addition to obtaining the same effects as the first embodiment, some maintenance parts are excluded from the inventory allocation plan generation process of this invention, making it possible to select a more appropriate inventory allocation plan.

[0063] The present invention is not limited to the embodiments described above, and includes various modifications. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations. [Explanation of Symbols]

[0064] 100...Inventory placement plan determination device, NW...Network, 200...User terminal, 300...Database, 110...Storage unit, 120...Calculation unit, 130...Input unit, 140...Output unit, 111...Item master information storage unit, 112...Demand information storage unit, 113...Inventory information storage unit, 114...Inventory placement plan information storage unit, 115...Search history information storage unit, 116...Machine operation information storage unit, 117...Prediction result information storage unit, 118...Evaluation index calculation result information storage unit, 121...Demand occurrence timing prediction unit, 122...Inventory placement plan generation unit, 123...Evaluation index calculation unit, 1400...Output screen.

Claims

1. In an inventory placement plan determination device having a control device and a storage device, which determines the placement of parts inventory at a logistics base, The aforementioned storage device is Equipped with a memory unit for storing various types of information, The control device is The system includes a calculation unit that performs various calculations related to the arrangement of the inventory of the parts based on the information stored in the storage unit. The storage unit stores at least demand information regarding the demand for the parts and search history information regarding the search history for the parts. The aforementioned arithmetic unit, A demand timing prediction unit that predicts the timing at which demand for the part will occur in the future, based at least on the demand information and the search history information, Based on the forecast results from the demand timing forecasting unit, an inventory allocation plan generation unit generates an inventory allocation plan for the inventory of parts at the logistics base. An inventory placement plan determination device characterized by comprising the following features.

2. The inventory placement plan determination device according to claim 1, further comprising an evaluation index calculation unit that calculates the service rate of the logistics base and the inventory value based on the inventory placement plan generated by the inventory placement plan generation unit and the timing at which demand for the parts is predicted by the demand timing prediction unit.

3. The inventory placement plan determination device according to claim 1, wherein the demand timing prediction unit predicts the timing at which demand for the part occurs using the causal relationship between the operating information of the machine in which the part is used and the search history information.

4. The memory unit further stores component characteristic information indicating the characteristics of the component, The inventory placement plan determination device according to claim 1, wherein the demand timing prediction unit predicts the timing based on the demand information, the search history information, and the component characteristic information.

5. The inventory placement plan determination device according to claim 4, wherein the demand generation timing prediction unit stops predicting the timing according to the characteristics of the component, and the inventory placement plan generation unit stops generating the inventory placement plan.

Images