Inventory Management Scheduling Optimization System and Method Based on Data Prediction

Abstract

The present invention discloses an inventory management scheduling optimization system and method based on data prediction, which relates to the technical field of inventory management and is used to improve problems such as insufficient or excessive equipment inventory caused by changes in the equipment iteration speed. It includes identifying equipment models to determine equipment information, retrieving equipment based on the equipment information, obtaining historical data of the same type of equipment, using recursive partitioning to delimit equipment categories according to the historical data, setting an analysis time, obtaining the equipment transfer volume within the analysis time, setting the sensitive time level of the equipment based on the comprehensive equipment transfer volume and equipment category, formulating an inventory retrieval mode according to the sensitive time level of the equipment, calculating the change prediction coefficient of equipment movement according to the inventory retrieval mode, and simultaneously monitoring the release information of new equipment of the same type. By comprehensively using the release information of new equipment of the same type and the change prediction coefficient, the Robust Scaling algorithm and the decision function are used to change the inventory retrieval mode, thereby reducing the problems of insufficient or excessive inventory caused by the iteration rates of different equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of inventory management. More specifically, the present invention relates to an inventory management scheduling optimization system and method based on data prediction. Background Art

[0002] Inventory management technology is an important tool to ensure optimized inventory levels, reduce costs, and improve service levels. The application of inventory management technology in electronic equipment warehouses improves the management efficiency of electronic equipment and provides assistance for formulating the transfer plan of electronic equipment.

[0003] The prior art has the following deficiencies:

[0004] In the past, when calling electronic equipment in inventory, by setting a storage threshold in advance, the inventory in the warehouse will be retrieved only when the in-store inventory is lower than the storage threshold. However, a large retrieval quantity will increase the retrieval time. As a technology product, electronic equipment has a fast update and iteration speed, and different products have different update and iteration speeds. Under a fixed storage threshold, different product iteration speeds will lead to multiple problems such as excessive inventory and insufficient inventory. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an inventory management scheduling optimization system and method based on data prediction. By analyzing and classifying electronic equipment and analyzing the sensitive time of the marked equipment, a sensitive time inspection mechanism is set to select different retrieval modes to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An inventory management scheduling optimization method based on data prediction, comprising the following steps:

[0008] Step S1: Identify the device model to determine device information, retrieve the historical data of the same type of device according to the device information, and delimit the device category according to the historical data;

[0009] Step S2: Set the analysis time, obtain the device transfer quantity, and set the sensitive time level by integrating the device transfer quantity and the device category;

[0010] Step S3: Select the inventory retrieval mode according to the sensitive time level. When the sensitive time of the device is low, collect the proportion of the device name entries and calculate the change prediction coefficient based on the warehouse retrieval quantity;

[0011] Step S4: Monitor the release information of new devices of the same type, and change the inventory retrieval mode by integrating the release information of new devices of the same type and the change prediction coefficient.

[0012] In a preferred embodiment, in step S1, the information tag of the electronic device is identified through the RFID system to obtain the device name and production location. Devices with the same name prefix are classified as devices of the same series. The device name prefix is retrieved in the website corresponding to the production location to obtain the number of devices in the same series and the release time interval of the devices in the same series. The sum of the release time intervals of the devices in the same series is averaged to obtain the series update value of the corresponding electronic device;

[0013] The RFID system is a wireless communication technology that identifies electronic devices through radio frequency signals. The website of the production location of the electronic device is used to provide information related to the manufacture of the electronic device, including the device name prefix.

[0014] In a preferred embodiment, in step S1, the series update values of different series of devices of the same type are obtained through the website of the production location of the electronic device. The device classification threshold is set using recursive segmentation according to the series update values of different series of devices. The specific steps are as follows:

[0015] Arrange the series update values of different series of devices from smallest to largest. Set the initial segmentation score and the upper segmentation score. Use the initial segmentation score to segment the sequence, divide the sequence into two parts, calculate the means of the two parts, compare the two parts of the sequence, select the sequence with the larger mean, reduce the denominator of the segmentation score by a preset reduction coefficient to increase the value of the segmentation score, use the segmentation score to perform the same segmentation on the remaining sequence, select the sequence with the smaller mean, and repeat the operation until the segmentation score reaches the upper segmentation score. Take the average value of the values in the remaining sequence as the device classification threshold;

[0016] Different series of devices are devices with different device name prefixes.

[0017] In a preferred embodiment, in step S1, the series update value of the corresponding electronic device is compared with the calculated device classification threshold to classify the electronic device. When the series update value of the electronic device exceeds the device classification threshold, the electronic device is classified as a low-speed iteration device; when the series update value of the electronic device is lower than the device classification threshold, the electronic device is classified as a high-speed iteration device.

[0018] In a preferred embodiment, in step S2, a period of time is selected as the analysis time to mark the electronic device. The ratio of the device transfer volume to the total inventory of the marked device is used as the transfer ratio. The transfer ratios of the devices outside the electronic device series where the marked device is located are calculated, and their average value is used as the transfer threshold;

[0019] When the marked device is a high-speed iteration device, the iteration coefficient of the marked device is set to 1; when the marked device is a low-speed iteration device, the iteration coefficient of the marked device is set to 0;

[0020] Set the sensitive time level rule according to the transfer ratio, transfer threshold and iteration coefficient of the marking device. The specific steps are as follows:

[0021] Rule 1: Add the iteration coefficient of the marking device to the transfer ratio to obtain the sensitivity judgment index, and add 1 to the transfer threshold as the sensitivity judgment threshold.

[0022] Rule 2: When the sensitivity judgment index of the marking device exceeds the sensitivity judgment threshold, it is judged that the sensitive time level of the marking device is high; when the sensitivity judgment index of the marking device is lower than the sensitivity judgment threshold, it is judged that the sensitive time level of the marking device is low.

[0023] In a preferred embodiment, in step S3, detect the warehouse transfer speed of the marking device and calculate the timeliness coefficient of the marking device, and determine the device change prediction coefficient by integrating the warehouse transfer speed and the timeliness coefficient of the marking device;

[0024] Collect the search ratio of the name entries of the marking device, use the ratio of the name entries of the marking device to the total number of search entries as the actual timeliness value of the marking device, use the ratio of the category to which the marking device belongs to the total number of categories of all electronic devices in the production area as the predicted timeliness value of the marking device, and use the ratio of the actual timeliness value of the marking device to the predicted timeliness value as the timeliness coefficient of the marking device;

[0025] Record the average value of the warehouse transfer quantities of all electronic devices in the production area, and use the ratio of the warehouse transfer quantity of the marking device to the calculated average value as the warehouse transfer speed of the marking device.

[0026] In a preferred embodiment, in step S3, calculate the warehouse transfer speed of all electronic devices in the production area, and calculate the change prediction coefficient using the Robust Scaling algorithm. The specific steps are as follows:

[0027] Merge the warehouse transfer speeds of each electronic device into a speed regulation data set, and calculate the warehouse transfer rate of the marking device using the Robust Scaling formula: , where y is the warehouse transfer rate of the marking device, x is the warehouse transfer speed of the marking device, is the average value in the speed regulation data set, is the difference between the 1 / 4 quantile and the 3 / 4 quantile when arranged from small to large in the speed regulation data set; add the warehouse transfer rate of the marking device to the timeliness coefficient to obtain the change prediction coefficient.

[0028] In a preferred embodiment, in step S4, monitor the release information of new devices of the same type of the marking device, record the release date, calculate the difference between the monitoring date and the release date to obtain the iteration time difference, calculate the difference between the monitoring date and the release date of the marking device to obtain the scheduling time difference, use the ratio of the scheduling time difference to the iteration time difference as the iteration coefficient, and use the decision function to change the inventory retrieval rule by combining the iteration coefficient and the change prediction coefficient. The specific steps are as follows:

[0029] Calculate the decision value of the marking device through the decision function using the obtained iteration coefficient and change prediction coefficient of the marking device. The formula is: , where x is the change prediction coefficient of the marking device, is the decision value of the marking device, is the iteration coefficient of the marking device, and T and b are the iteration coefficient modulation amount and error amount respectively;

[0030] When the calculated decision value of the marking device exceeds the preset decision threshold, it is determined that the marking device needs to perform the standard retrieval mode. Otherwise, it is determined that the marking device performs the simple retrieval mode;

[0031] The simple retrieval mode is to regularly retrieve devices using the preset retrieval quantity;

[0032] In the standard retrieval mode, the preset retrieval quantity is used as the adjusted retrieval quantity of the marking device by multiplying the warehouse retrieval speed by the preset retrieval quantity to retrieve the device.

[0033] The inventory management scheduling optimization system based on data prediction is used to implement the above-mentioned inventory management scheduling optimization method based on data prediction, and includes a data collection module, a sensitivity determination module, an information monitoring module, and a retrieval processing module;

[0034] The data collection module is used to determine device information by the device model, collect historical data of the same type of devices and the device transfer quantity and transmit them to the sensitivity determination module;

[0035] The sensitivity determination module is used to set the sensitivity time level for the marking device and calculate the change prediction coefficient and transmit it to the retrieval processing module;

[0036] The information monitoring module is used to detect the release information of new devices of the same type and calculate the iteration coefficient of the marking device and transmit it to the retrieval processing module;

[0037] The retrieval processing module is used to select to process the marking device in the simple retrieval mode or the standard retrieval mode.

[0038] The technical effects and advantages of the inventory management scheduling optimization system and method based on data prediction of the present invention:

[0039] The present invention determines device information by identifying the device model, retrieves the device based on the device information, obtains the historical data of the same type of device, delimits the device category according to the historical data, separates and processes the devices by delimiting the device category, narrows the selection direction of the retrieval mode, reduces the problem of device inventory failure, sets the analysis time, obtains the device replenishment quantity within the analysis time, sets the sensitive time level of the device based on the comprehensive device replenishment quantity and device category, formulates the inventory retrieval mode according to the sensitive time level of the device, calculates the change prediction coefficient of the device movement according to the inventory retrieval mode, monitors the release information of the same type of new device at the same time, and changes the inventory retrieval mode by comprehensively considering the release information of the same type of new device and the change prediction coefficient, so as to reduce the problem of inventory shortage or overstock caused by the iteration rate of different devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a schematic diagram of the inventory management scheduling optimization method based on data prediction of the present invention,

[0041] Figure 2 is a flowchart of the inventory management scheduling optimization system based on data prediction of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] The present invention determines device information by identifying the device model, retrieves the device based on the device information, obtains the historical data of the same type of device, delimits the device category according to the historical data, sets the analysis time, obtains the device replenishment quantity within the analysis time, sets the sensitive time level of the device based on the comprehensive device replenishment quantity and device category, formulates the inventory retrieval mode according to the sensitive time level of the device, calculates the change prediction coefficient of the device movement according to the inventory retrieval mode, monitors the release information of the same type of new device at the same time, and changes the inventory retrieval mode by comprehensively considering the release information of the same type of new device and the change prediction coefficient, so as to reduce the problem of inventory shortage or overstock caused by the iteration rate of different devices.

[0044] Embodiment 1, an inventory management scheduling optimization method based on data prediction, as Figure 1 shown, includes the following steps:

[0045] Step S1: Identify the device model to determine the device information, retrieve the historical data of the same type of device according to the device information, and delimit the device category according to the historical data;

[0046] Step S2: Set the analysis time, obtain the equipment transfer volume, and set the sensitive time level based on the comprehensive equipment transfer volume and equipment category;

[0047] Step S3: Develop an inventory retrieval mode according to the sensitive time level, and calculate the change prediction coefficient according to the inventory retrieval mode;

[0048] Step S4: Monitor the release information of new equipment of the same type, and change the inventory retrieval mode based on the comprehensive release information of new equipment of the same type and the change prediction coefficient.

[0049] The specific implementation is as follows:

[0050] In step S1, when the electronic device is produced, an information tag will be generated. The information tag contains the basic information of the electronic device, including the device name, production location, warranty period, etc. When the electronic device enters the warehouse, the electronic device is identified through the RFID system to determine the type, brand of the electronic device, and the device series in the production company where the electronic device is located. The type of the electronic device can use a pre-set device name tag. For example, the types of electronic devices include laptop computers, smart furniture, etc.

[0051] The device series is the same series of devices of the identified electronic device in the corresponding production company. Usually, the production company will keep the name prefix of the same series of devices consistent and add a digital serial number to the name suffix to identify different generations of products in the same series. Therefore, devices with the same name prefix can be classified as the same series of devices.

[0052] The historical data is the release time interval of the same series of devices. The information tag of the electronic device is identified through the RFID system to obtain the basic information of the electronic device, confirm the device name and production location of the electronic device, access the website of the electronic device production location to retrieve the device name prefix, obtain the number of devices in the same series and the release time interval of the same series of devices, and sum and average the release time intervals of the same series of devices to obtain the series update value of the corresponding electronic device.

[0053] Use the same method to calculate the series update values of devices of the same type but different series in the same production location. According to the series update values of different series of devices, use recursive segmentation to set the device classification threshold. The specific steps are as follows:

[0054] Arrange the series update values of different series of devices from small to large, set the initial segmentation score and the upper limit segmentation score, use the initial segmentation score to segment the sequence, discard the sequence with a smaller mean value and a smaller number of values, reduce the denominator of the segmentation score by a preset reduction coefficient to increase the value of the segmentation score, use the segmentation score to perform the same segmentation on the remaining sequence, discard the sequence with a larger mean value and a smaller number of values, repeat the operation until the segmentation score reaches the upper limit segmentation score, and take the average value of the values in the remaining sequence as the device classification threshold.

[0055] For example, there are 8 series of the same type of devices in the same production location. The series update values of the 8 series of devices are calculated, arranged from smallest to largest. The initial segmentation score is set to 1 / 4, and the upper limit segmentation score is set to 1 / 2. The series update value sequence is divided into two parts according to 1 / 4, the means of the two parts are calculated, the sequence with the smaller mean and fewer numerical quantities is discarded, and the increase coefficient is 1. Then, after sorting the series update values of the remaining 6 series of devices from smallest to largest, the segmentation score becomes 1 / 3. The series update value sequence is segmented according to 1 / 3, and the sequence with the larger mean and fewer numerical quantities is discarded. Repeat the operation until the segmentation score reaches the upper limit segmentation score, and calculate the average value among the remaining values as the device classification threshold.

[0056] Compare the series update value of the corresponding electronic device with the calculated device classification threshold to classify the electronic device. When the series update value of the electronic device exceeds the device classification threshold, the electronic device is classified as a low-speed iteration device; when the series update value of the electronic device is lower than the device classification threshold, the electronic device is classified as a high-speed iteration device.

[0057] It should be noted that the RFID system is a wireless communication technology used to identify and track objects through radio frequency signals, which can be used to identify electronic devices. The electronic device production site website is used to provide information and services related to electronic product manufacturing, and can obtain the types and models of electronic devices produced in the corresponding production location and the release times of different types or models of electronic devices. The series update value of an electronic device reflects the iteration rate of the corresponding electronic device. The higher the iteration rate, the shorter the sensitive time for inventory placement. By using the recursive segmentation method to set the device classification threshold, the requirements or accuracy of device classification can be adjusted by adjusting the upper limit segmentation threshold and reducing the coefficient, providing a data basis for setting the sensitive time level in the future.

[0058] In step S2, select a period of time as the analysis time. During the analysis time, collect the device transfer volume of the electronic device, mark the electronic device, calculate the sold ratio according to the device transfer volume and the total inventory, use the ratio of the device transfer volume to the total inventory as the transfer ratio, and use the same method to calculate the transfer ratio of the electronic devices other than the types of electronic devices produced at the marked electronic device production location, and take the average value as the transfer threshold.

[0059] When the marked device is a high-speed iteration device, set the iteration coefficient of the marked device to 1; when the marked device is a low-speed iteration device, set the iteration coefficient of the marked device to 0.

[0060] Set the sensitive time level rules according to the transfer ratio, transfer threshold, and iteration coefficient of the marked device. The specific steps are as follows:

[0061] Rule 1: Sum the iteration coefficient of the marking device and the transfer ratio to obtain the sensitivity judgment index, and sum the transfer threshold and 1 as the sensitivity judgment threshold.

[0062] Rule 2: When the sensitivity judgment index of the marking device exceeds the sensitivity judgment threshold, it is determined that the sensitive time level of the marking device is high; when the sensitivity judgment index of the marking device is lower than the sensitivity judgment threshold, it is determined that the sensitive time level of the marking device is low.

[0063] It should be noted that the sensitive time is the high-frequency time of the electronic device in warehouse scheduling. The sensitive time is determined by the iteration rate and transfer ratio of the electronic device. When the same type of electronic device undergoes iteration, the scheduling rate in the warehouse will decrease; if the sensitive time is long, the scheduling rate of the electronic device will be at a high level for a long time, and if the sensitive time is short, the scheduling rate of the electronic device will be at a low level for a long time.

[0064] In step S3, when the sensitive time of the marking device is high, the simple retrieval mode is performed. When the sensitive time of the marking device is low, the sensitive time inspection mechanism is entered, and after inspecting the sensitive time of the marking device, it is determined whether to perform the standard retrieval mode.

[0065] When the sensitive time level of the marking device is high, that is, the high-frequency time for warehouse scheduling is sufficient, and the tolerance for reserving and dispatching the marking devices in stock is higher. In the simple retrieval mode, goods are dispatched and the warehouse is replenished through a preset time interval.

[0066] When the sensitive time level of the marking device is low, that is, the high-frequency time for warehouse scheduling is insufficient, and the tolerance for reserving and dispatching the marking devices in stock is low. It is necessary to perform real-time detection and adjustment of the transfer behavior to balance the supply and demand of the marking devices.

[0067] Before processing the marking device in the standard retrieval mode, detect the warehouse retrieval speed of the marking device and calculate the timeliness coefficient of the marking device. Adjust the preset quantity of goods dispatched and warehouse replenishment based on the comprehensive warehouse retrieval speed and timeliness coefficient of the marking device.

[0068] When calculating the timeliness coefficient of the marking device, by recording the search ratio of the name entries in the website of the production place of the electronic device of the marking device, taking the ratio of the name entries of the marking device to the total number of search entries recorded in the website database as the actual timeliness value of the marking device, taking the ratio of the category type of the marking device to the total number of category types of all electronic devices in the website database as the predicted timeliness value of the marking device, and taking the ratio of the actual timeliness value of the marking device to the predicted timeliness value as the timeliness coefficient of the marking device.

[0069] When the aging coefficient of the marking device exceeds 1, that is, the actual aging value of the marking device is greater than the predicted aging value, it is judged that the marking device is still in the aging period, and the retrieval is still at a high frequency level; when the aging coefficient of the marking device is lower than 1, that is, the actual aging value of the marking device is lower than the predicted aging value, it is judged that the marking device is outside the aging period, and the retrieval level drops.

[0070] Select a time interval to record the number of retrievals from the warehouse of the marking device. Calculate the average value of the number of retrievals from the warehouse of all electronic devices in the website database within the same time interval. Take the ratio of the number of retrievals from the warehouse of the marking device to the calculated average value as the retrieval speed of the marking device. Multiply the retrieval speed by the product of the preset initial goods out and the number of warehouse replenishments as the adjusted number of goods out and warehouse replenishments of the marking device.

[0071] Use the same method to calculate the retrieval speed of all electronic devices, and use the Robust Scaling algorithm to calculate the change prediction coefficient. The specific steps are as follows:

[0072] Merge the retrieval speeds of each electronic device into a speed adjustment data set, and use the Robust Scaling formula to calculate the warehouse adjustment rate of the marking device: , where y is the warehouse adjustment rate of the marking device, x is the retrieval speed of the marking device, is the average value in the speed adjustment data set, is the difference between the 1 / 4 quantile and the 3 / 4 quantile when arranged from small to large in the speed adjustment data set. Sum the warehouse adjustment rate of the marking device and the aging coefficient to obtain the change prediction coefficient.

[0073] It should be noted that the Robust Scaling algorithm is a data standardization algorithm. In this example, by compressing the retrieval speed of the marking device and calculating the change prediction coefficient, it is more convenient to regulate the subsequent change of the inventory retrieval mode.

[0074] In step S4, monitor the release information of new devices of the same type as the marking device, record the release date, subtract the release date from the monitoring date to obtain the iteration time difference, subtract the release date of the marking device from the monitoring date to obtain the scheduling time difference, and take the ratio of the scheduling time difference to the iteration time difference as the iteration coefficient. Use the decision function to perform sensitive time level modulation by combining the iteration coefficient and the change prediction coefficient, and change the inventory retrieval rule. The specific steps are as follows:

[0075] Calculate the decision value of the marking device through the decision function using the calculated iteration coefficient and change prediction coefficient. The formula is: , where x is the change prediction coefficient of the marking device, is the decision value of the marking device, Let \(k\) be the iteration coefficient of the marking device, \(T\) and \(b\) be the modulation amount and error amount of the iteration coefficient respectively. The larger the iteration coefficient or the larger the variation prediction coefficient, the larger the decision value of the marking device, and the more necessary it is to perform the standard retrieval mode.

[0076] By calculating the decision value of the marking device, when the decision value of the marking device obtained by calculation exceeds the preset decision threshold, it is determined that the marking device needs to perform the standard retrieval mode; otherwise, it is determined that the marking device performs the simple retrieval mode.

[0077] It should be noted that the modulation amount and error amount of the iteration coefficient and the decision threshold can be set through experiments or following the opinions of professionals in the field, and will not be analyzed here.

[0078] An inventory management scheduling optimization system based on data prediction, as Figure 2 shown, is used to implement an inventory management scheduling optimization method based on data prediction, including a data acquisition module, a sensitivity determination module, an information monitoring module, and a retrieval processing module;

[0079] The data acquisition module is used to determine device information based on the device model, and collect historical data of the same type of device and the device transfer volume and transmit them to the sensitivity determination module;

[0080] The sensitivity determination module is used to set the sensitive time level for the marking device and calculate the variation prediction coefficient and transmit it to the retrieval processing module;

[0081] The information monitoring module is used to detect the release information of new devices of the same type, and calculate the iteration coefficient of the marking device and transmit it to the retrieval processing module;

[0082] The retrieval processing module is used to select to perform simple retrieval mode processing or standard retrieval mode processing on the marking device.

[0083] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0084] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application of the technical solution and the inventive constraints. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0085] In addition, in each embodiment of the present application, each functional module may be integrated into one processing module, may exist separately as individual physical modules, or two or more modules may be integrated into one module.

[0086] As described above, the foregoing are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0087] Finally, the foregoing are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. The inventory management scheduling optimization method based on data prediction is characterized by: The following steps are included: Step S1: Identify the device model to determine the device information, retrieve the historical data of the same type of device according to the device information, and define the device category according to the historical data; Step S2: Set the analysis time, obtain the equipment transfer quantity, and set the sensitive time level based on the equipment transfer quantity and equipment category; Step S3: Select the inventory retrieval mode according to the sensitive time level. When the sensitive time of the equipment is low, collect the equipment name entry ratio and the warehouse retrieval quantity to calculate the change prediction coefficient; Step S4: monitor the release information of new equipment of the same type, and change the inventory retrieval mode based on the release information of new equipment of the same type and the change prediction coefficient; The inventory retrieval mode includes a simple retrieval mode and a standard retrieval mode; In step S3, the warehouse retrieval speed of the marking equipment is detected and the timeliness coefficient of the marking equipment is calculated, and the change prediction coefficient of the equipment is determined by combining the warehouse retrieval speed of the marking equipment and the timeliness coefficient; Collect the search ratio of the name terms of the marked device, take the ratio of the name terms of the marked device to the total number of search terms as the actual timeliness value of the marked device, take the ratio of the category to which the marked device belongs to the total number of categories of all electronic equipment at the production site as the predicted timeliness value of the marked device, and take the ratio of the actual timeliness value of the marked device to the predicted timeliness value as the timeliness coefficient of the marked device; Record the warehouse retrieval quantity of all electronic equipment at the production site and calculate the average value, and use the ratio of the warehouse retrieval quantity of the marking equipment to the calculated average value as the warehouse retrieval speed of the marking equipment; In step S3, the warehouse retrieval speed of all electronic devices at the production site is calculated, and the change prediction coefficient is calculated using the Robust Scaling algorithm. The specific steps are as follows: The warehouse retrieval speed of each electronic device is combined into a speed regulation data set, and the warehouse regulation rate of the marked device is calculated using the Robust Scaling formula: , where y is the warehouse adjustment rate of the marking device, and x is the warehouse retrieval speed of the marking device. is the average value in the speed regulation data set, It is the difference between the 1 / 4 quantile and the 3 / 4 quantile when the speed regulation data set is arranged from small to large; the change prediction coefficient is obtained by summing the warehouse regulation rate and the time efficiency coefficient of the marked equipment.

2. The inventory management scheduling optimization method based on data prediction according to claim 1 is characterized by: In step S1, the information tag of the electronic device is identified by the RFID system to obtain the device name and production place, and the devices with the same name prefix are classified as the same series of devices. The device name prefix is ​​searched on the website of the corresponding production place to obtain the number of devices in the same series and the release time interval of the devices in the same series. The release time intervals of the devices in the same series are summed and averaged to obtain the series update value of the corresponding electronic device; The RFID system is a wireless communication technology that identifies electronic devices through radio frequency signals. The website where the electronic device is produced is used to provide information related to the manufacture of the electronic device, including the device name prefix.

3. The inventory management scheduling optimization method based on data prediction according to claim 2 is characterized by: In step S1, the series update values ​​of different series of devices of the same type are obtained through the website of the electronic device manufacturer, and the device classification threshold is set by recursive segmentation according to the series update values ​​of different series of devices. The specific steps are as follows: Arrange the series update values ​​of different series of equipment from small to large, set the initial segmentation score and the upper limit segmentation score, use the initial segmentation score to segment the sequence, discard the sequence with a small mean and a small number of values, reduce the denominator of the segmentation score by a preset reduction coefficient to increase the value of the segmentation score, use the segmentation score to perform the same segmentation on the remaining sequences, discard the sequence with a large mean and a small number of values, repeat the operation until the segmentation score reaches the upper limit segmentation score, and take the average value of the values ​​in the remaining sequences as the device classification threshold; Devices of different series have different prefixes in their device names.

4. The inventory management scheduling optimization method based on data prediction according to claim 2 is characterized by: In step S1, the series update value of the corresponding electronic device is compared with the calculated device classification threshold to classify the electronic device. When the series update value of the electronic device exceeds the device classification threshold, the electronic device is classified as a low-speed iteration device; when the series update value of the electronic device is lower than the device classification threshold, the electronic device is classified as a high-speed iteration device.

5. The inventory management scheduling optimization method based on data prediction according to claim 4 is characterized by: In step S2, a period of time is selected as the analysis time to mark the electronic devices, the ratio of the equipment transfer quantity of the marked device to the total inventory is used as the transfer ratio, the equipment transfer ratio outside the electronic equipment series where the marked device belongs is calculated, and the average value is taken as the transfer threshold; When the marking device is a high-speed iteration device, the iteration coefficient of the marking device is set to 1; when the marking device is a low-speed iteration device, the iteration coefficient of the marking device is set to 0; Set the sensitive time level rules according to the mobilization ratio, mobilization threshold and iteration coefficient of the marked device. The specific steps are as follows: Rule 1: The iteration coefficient of the marking device and the mobilization ratio are summed to obtain the sensitivity judgment index, and the mobilization threshold and 1 are summed to obtain the sensitivity judgment threshold; Rule 2: When the sensitivity judgment index of the marking device exceeds the sensitivity judgment threshold, the sensitive time level of the marking device is judged to be high; when the sensitivity judgment index of the marking device is lower than the sensitivity judgment threshold, the sensitive time level of the marking device is judged to be low.

6. The inventory management scheduling optimization method based on data prediction according to claim 1 is characterized by: In step S4, the release information of new devices of the same type as the marked device is monitored, the release date is recorded, the monitoring date is subtracted from the release date to obtain the iteration time difference, the monitoring date is subtracted from the release date of the marked device to obtain the scheduling time difference, the ratio of the scheduling time difference to the iteration time difference is used as the iteration coefficient, and the decision function is used to change the inventory retrieval rule based on the iteration coefficient and the change prediction coefficient. The specific steps are as follows: The iteration coefficient and change prediction coefficient calculated by the marking device are used to calculate the decision value through the decision function, and the formula is: , where x is the change prediction coefficient of the marking device, is the decision value of the marking device, is the iteration coefficient of the marking device, T and b are the iteration coefficient modulation and error respectively; When the calculated decision value of the marking device exceeds the preset decision threshold, it is determined that the marking device needs to be in the standard call mode, otherwise, it is determined that the marking device is in the simple call mode; The simple retrieval mode is to regularly use the preset retrieval volume to retrieve the device; The standard retrieval mode uses the preset retrieval quantity to retrieve equipment by multiplying the warehouse retrieval speed by the preset retrieval quantity as the adjusted retrieval quantity for the marked equipment.

7. An inventory management scheduling optimization system based on data prediction, based on the inventory management scheduling optimization method based on data prediction according to any one of claims 1 to 6, characterized in that: It includes data collection module, sensitivity determination module, information monitoring module and retrieval processing module; The data collection module is used to determine the equipment information by equipment model, collect historical data of the same type of equipment, and transfer the equipment transfer quantity to the sensitive judgment module; The sensitive determination module is used to set the sensitive time level for the marking device and calculate the change prediction coefficient to be transmitted to the retrieval processing module; The information monitoring module is used to detect the release information of new devices of the same type, and calculate the iteration coefficient of the marked device and pass it to the call processing module; The retrieval processing module is used to select simple retrieval mode processing or standard retrieval mode processing for the marking device.

Images