Supply chain management method and system based on digitization

Abstract

The invention relates to the technical field of data processing, and particularly discloses a digitalization-based supply chain management method and system, and the method comprises the steps: obtaining the historical order quantity of goods; clustering the historical order quantity by using an ordered sample clustering method to obtain a plurality of order quantity sections; obtaining a peak season center section and a slack season center section in the plurality of order quantity sections, calculating a peak season merging coefficient and a slack season merging coefficient of any remaining order quantity section, and performing merging to obtain a peak season section set, a slack season section set and a non-section set; and obtaining the sub-sections of each order quantity section in the peak season section set, the slack season section set and the non-section set with the highest correlation degree with the order quantity of the current time period, so as to judge the historical order quantity configuration supply quantity after the time period in which the corresponding sub-section is located when the coefficient is greater than a preset threshold value. According to the digitalization-based supply chain management method and system provided by the invention, the peak season and the slack season can be reasonably divided, and the problem of supply shortage or excess supply is avoided.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a digital-based supply chain management method and system. Background Art

[0002] Supply chain management has a crucial impact on the relationship between supply and demand. Through precise demand forecasting and response mechanisms, it helps companies rationally manage production and inventory, avoiding oversupply or shortages. It also strengthens information sharing and resource collaboration across regions, effectively balancing supply and demand and mitigating risk. It ensures timely product delivery, reduces inventory overstocks, and enhances customer trust and loyalty. By optimizing the supply-demand relationship, improving business efficiency and inventory management capabilities, supply chain management drives long-term growth and market competitiveness.

[0003] For example, in the Chinese patent application document with application publication number CN113743994A, a method, system, device and storage medium for supplier peak season forecasting is disclosed, wherein the method includes the following steps: obtaining historical data of the supplier; preprocessing and screening the historical data to obtain valid historical data; constructing a peak season forecasting model based on the seasonal autoregressive differential moving average algorithm and the valid historical data; obtaining monthly historical data from the valid historical data, inputting the monthly historical data into the peak season forecasting model, and obtaining the supplier's predicted monthly data for a period of time in the future; marking the peak season in the predicted monthly data to obtain the supplier's peak season for a period of time in the future. However, the seasonal autoregressive differential moving average algorithm in the above scheme is highly complex, has many parameters, is difficult to adjust, and requires determining the seasonal cycle, making it inconvenient to calculate in the actual forecasting process.

[0004] The ordered sample clustering method divides ordered samples into clusters by finding the optimal split point, minimizing sample variance within each cluster and maximizing inter-cluster variance. Therefore, it can be used to cluster historical supplier data, identifying peak and off-season segments. Supply data from these peak and off-season segments can then be used as a reference for supply in the current segment.

[0005] However, the ordered sample clustering algorithm cannot effectively distinguish the transition segment between the peak season segment and the off-season segment. This is because it divides the transition segment into multiple segments, thereby shortening the duration of the peak season segment and the off-season segment. Using this as a reference will cause supply shortages or oversupply problems, and will cause companies to frequently adjust supply, resulting in increased transportation costs and waste of resources. Summary of the Invention

[0006] The present invention provides a digital-based supply chain management method and system, aiming to solve the technical problem in the existing technology that it is impossible to effectively distinguish the transition period between the peak season and the off-season, resulting in supply shortage or oversupply when stocking up based on the peak season and the off-season.

[0007] A digital supply chain management method of the present invention comprises the following steps:

[0008] Get the historical order volume of goods;

[0009] Use the ordered sample clustering method to cluster the historical order volume and obtain multiple order volume segments;

[0010] Obtain peak season center segments and off-season center segments from multiple order volume segments, calculate the peak season merging coefficient and off-season merging coefficient of any segment in the remaining order volume segments, and merge any segment with the peak season center segment or the off-season center segment to obtain a peak season segment set and an off-season segment set; the remaining unmerged order volume segments are a non-segment set; wherein the merging coefficient is the product of the degree of proximity and stability between the order volume segment and the target center segment; the stability is the difference between 1 and the ratio of the duration between the order volume segment and the nearest target center segments on its left and right sides; the target center segment is the peak season center segment or the off-season center segment;

[0011] The sub-segments of each order volume segment in the peak season segment set, off-season segment set and non-segment set that have the highest degree of correlation with the order volume of the current time period are obtained respectively, and the supply quantity is configured according to the historical order volume after the time period of the corresponding sub-segment when the judgment coefficient is greater than the preset threshold; wherein the judgment coefficient is the product of the ratio of the current time period length and the order volume segment length of the corresponding sub-segment and the degree of proximity; the degree of proximity represents the similarity between the order volume of the current time period and the order volume in the sub-segment.

[0012] In the above scheme, the order volume segments are merged into the corresponding peak season central segments and off-season central segments according to the merging coefficient of the peak season and the merging coefficient of the off-season, thereby obtaining the peak season segment set, the off-season segment set and the flat season segment set, so that the division of the peak season, the off-season and the flat season is more accurate and more reasonable, and according to the degree of correlation and the judgment coefficient, the sub-segment of the peak season segment set, the off-season segment set or the flat season segment set that is most similar to the order volume of the current period can be obtained, and the supply volume is configured according to the historical order volume after the period where the corresponding sub-segment is located, so that reasonable stocking can be achieved, avoiding the problem of supply shortage or oversupply caused by the inability to effectively distinguish the transition segment between the peak season and the off-season in the existing technology and stocking with the peak season and the off-season as a reference.

[0013] Preferably, the peak season central segment is the order volume segment corresponding to the maximum value of the order volume mean among all order volume segments, and the off-season central segment is the order volume segment corresponding to the minimum value of the order volume mean among all order volume segments.

[0014] In the above scheme, the corresponding peak season center segment and off-season center segment are obtained according to the extreme value of the mean order volume in all order volume segments, which makes the division of the peak season center segment and the off-season center segment more reasonable and facilitates the subsequent merging of the remaining order volume segments.

[0015] Preferably, the calculation of the peak season merging coefficient and the off-season merging coefficient of any segment in the remaining order volume segments to merge any segment with the peak season center segment or the off-season center segment includes:

[0016] When the peak season merging coefficient of any of the remaining order volume segments is greater than the preset merging coefficient, the order volume segment is merged into the peak season center segment;

[0017] When the off-season merging coefficient of any segment in the remaining order volume segments is greater than the preset merging coefficient, the order volume segment is merged into the off-season central segment.

[0018] Preferably, the proximity k1 between the order volume segment n and the peak season center segment a is:

[0019]

[0020] The degree of proximity k2 between the order volume segment n and the off-season center segment b is:

[0021]

[0022] Where, mean(n) is the mean order volume of order volume segment n, mean(a) is the mean order volume of the peak season center segment a closest to order volume segment n, and mean(b) is the mean order volume of the off-season center segment b closest to order volume segment n. The peak season center segment a and the off-season center segment b are located on the left and right sides of order volume segment n, respectively.

[0023] In the above scheme, the degree of proximity is calculated by comparing the average order volume within the order volume segment with the average order volume in the peak season center segment and the average order volume in the off-season center segment, which can truly reflect the degree of proximity of the order volume segment to the peak season center segment and the off-season center segment respectively.

[0024] Preferably, the degree of correlation is the Pearson correlation coefficient obtained by respectively calculating the order volume of the current period and the order volume of the sub-segments of each order volume segment in the peak season segment set, the off-season segment set and the non-segment set.

[0025] In the above scheme, the Pearson correlation coefficient is used to calculate the degree of correlation, which has the advantages of simple calculation, high efficiency, and wide application range.

[0026] Preferably, the sub-segments are obtained by moving a sliding window on each order volume segment in the peak season segment set, the off-season segment set and the non-segment set respectively; wherein the size of the sliding window is equal to the duration of the current period.

[0027] Preferably, the degree of proximity c is:

[0028]

[0029] Where, mean(Q i ) is the current period Q i The mean of the order quantity, mean(Q j ) is the sub-segment Q j The mean of the order quantity, and the sub-segment Q j With the current period Q i of equal length.

[0030] Preferably, configuring the supply quantity based on the historical order quantity after the period of the corresponding sub-segment where the judgment coefficient is greater than a preset threshold comprises:

[0031] Obtain the corresponding historical order volume after the period of the corresponding sub-segment of the current period in different regions, and use the corresponding historical order volume as the required supply volume for the corresponding region;

[0032] Sort the required supply quantities from largest to smallest, and allocate the required supply quantities for the corresponding regions in the order of the required supply quantities.

[0033] In the above scheme, the required supply quantities of different regions are allocated in sequence according to the order of the required supply quantities of different regions, so that the supply is more reasonable and the maximum economic benefit is achieved.

[0034] Preferably, the historical order volume is the historical order volume in days.

[0035] In the above solution, by obtaining the historical order volume in units of days, the obtained data is made more refined, more data details can be retained, and subsequent calculations are made more accurate.

[0036] The present invention also provides a digital-based supply chain management system, comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement any of the digital-based supply chain management methods described above.

[0037] The beneficial effects are:

[0038] The solution of the present invention merges the order volume segments into the corresponding peak season center segments and off-season center segments according to the merging coefficient of the peak season and the merging coefficient of the off-season, thereby obtaining the peak season segment set, the off-season segment set and the flat season segment set, so that the division of the peak season, the off-season and the flat season is more accurate and more reasonable, and according to the degree of correlation and the judgment coefficient, the sub-segment of the peak season segment set, the off-season segment set or the flat season segment set that is most similar to the order volume of the current time period can be obtained, and the supply volume is configured according to the historical order volume after the time period where the corresponding sub-segment is located, so that reasonable stocking can be achieved, avoiding the problem of supply shortage or oversupply caused by the inability to effectively distinguish the transition segment between the peak season and the off-season in the existing technology when stocking is carried out with the peak season and the off-season as a reference. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0040] Figure 1 A flowchart of the steps of a digital supply chain management method according to an embodiment of the present invention;

[0041] Figure 2 This is a structural block diagram of a digital supply chain management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0043] like Figure 1 According to a first aspect of the present invention, a digital supply chain management method is provided, comprising the following steps:

[0044] S1. Get the historical order volume of goods.

[0045] The goods of the present invention refer to goods with a sales cycle, that is, goods with a peak season and an off-season, such as seasonal fruits and vegetables, down jackets, cars and the like.

[0046] The historical order volume of goods can be obtained by reviewing the company's historical order records. Historical order volume is measured in daily units, that is, the number of goods ordered per day. This makes the data obtained more refined and retains more data details, thus making subsequent calculations more accurate.

[0047] The historical order volume in the present invention refers to a period of time, such as at least two years. Calculations, analyses, and forecasts based on at least two years of historical order volume provide more accurate forecasts due to the large amount of data and detailed analysis. This can mitigate errors caused by external factors or human factors.

[0048] S2. Use the ordered sample clustering method to cluster the historical order volume and obtain multiple order volume segments.

[0049] It should be noted that the ordered sample clustering method itself belongs to the existing technology, which requires that samples are arranged in a certain order and cannot be disrupted during classification, that is, samples of the same type must be adjacent to each other. The samples mentioned above are historical order quantities with time sequence, such as X1, X2, X3, ..., X n , where X n Indicates the historical order volume on day n.

[0050] The basic idea behind the ordered sample clustering method can be summarized as follows: order samples are divided into several classes by finding the optimal split point, which minimizes the variance within each class and maximizes the variance between classes. Specifically, the method first treats all samples as one large class, then gradually increases the number of classes, selecting the optimal split point based on a loss function each time the number of classes is increased. In this way, as the number of classes increases, the variance within a class decreases while the variance between classes increases, ultimately achieving an optimal classification result.

[0051] The ordered sample clustering method mainly includes the following core steps:

[0052] 1. Define the diameter of a class: The diameter of a class is an indicator to measure the difference within the class. The commonly used diameter is expressed as the sum of squares of the intra-class deviations;

[0053] 2. Obtain the loss function: The loss function is used to measure the quality of classification. The smaller the value, the more reasonable the classification. The commonly used loss function is the sum of the squared deviations within each class.

[0054] 3. Find the optimal classification: With the goal of minimizing the loss function, find the optimal segmentation point and obtain the optimal classification.

[0055] Therefore, multiple order volume segments can be obtained through the ordered sample clustering method, and each order volume segment includes multiple historical order volumes with time sequence.

[0056] However, when using the ordered sample clustering method to cluster historical order volumes, it divides historical order volumes into multiple order volume segments. Due to the excessive and overly detailed segmentation of the transition period between peak and off-season, the period that should have been in the peak or off-season is segmented out, thereby shortening the duration of the peak and off-season. Using this as a reference to configure supply volume can lead to supply shortages or oversupply, and companies will frequently adjust supply volume, resulting in increased transportation costs and wasted resources. Therefore, to avoid this situation, it is necessary to merge the periods that should have been in the peak or off-season during the transition period into the corresponding peak or off-season.

[0057] S3. Obtain peak season center segments and off-season center segments from the plurality of order volume segments, and calculate the peak season merging coefficient and off-season merging coefficient for any segment in the remaining order volume segments. Merge any segment with the peak season center segment or the off-season center segment to obtain a peak season segment set and an off-season segment set. The remaining unmerged order volume segments are referred to as non-segment sets.

[0058] Among them, the non-segment set is the off-season segment set, and the off-season segment set is the set that belongs neither to the peak season segment set nor to the off-season segment set.

[0059] Among them, the peak season center segment is the order volume segment corresponding to the maximum value of the order volume mean among all order volume segments, and the off-season center segment is the order volume segment corresponding to the minimum value of the order volume mean among all order volume segments.

[0060] This is because historical order volume refers to at least two years of historical order volume, and some products may have multiple peak and off-season periods within a year. Peak-season order volume segments tend to have higher order volumes and inevitably contain a segment with a corresponding maximum order volume mean, which is considered the peak-season center period. Similarly, off-season order volume segments tend to have lower order volumes and inevitably contain a segment with a corresponding minimum order volume mean, which is considered the off-season center period. Since there can be multiple maximum and minimum values, there are multiple peak-season and off-season center segments.

[0061] The above-mentioned calculation of the peak season merging coefficient and the off-season merging coefficient of any segment in the remaining order volume segment, so as to merge any segment with the peak season center segment or the off-season center segment, includes:

[0062] When the peak season merging coefficient of any of the remaining order volume segments is greater than the preset merging coefficient, the order volume segment is merged into the peak season center segment;

[0063] When the off-season merging coefficient of any segment in the remaining order volume segments is greater than the preset merging coefficient, the order volume segment is merged into the off-season central segment.

[0064] The preset merging coefficient has a value range of 0.6 to 0.8, preferably 0.7. Of course, the preset merging coefficient can be adjusted as needed.

[0065] The present invention introduces a merging coefficient to merge the corresponding order volume segments into the corresponding peak season center segment and the corresponding off-season center segment, respectively, to obtain multiple peak season segments centered on the peak season center segment and multiple off-season segments centered on the off-season center segment. The multiple peak season segments constitute a peak season segment set, and the multiple off-season segments constitute an off-season segment set.

[0066] The consolidation coefficient is the product of the proximity and stability of the order volume segment and the target center segment. Since the target center segment is either the peak season center segment or the off-season center segment, the peak season consolidation coefficient is the product of the proximity and stability of the order volume segment and the peak season center segment. The off-season consolidation coefficient is the product of the proximity and stability of the order volume segment and the off-season center segment.

[0067] The specific calculation process of the peak season and off-season combined coefficients is as follows:

[0068] First, the proximity between the order volume segment and the peak season center segment, as well as the proximity between the order volume segment and the off-season center segment, are obtained respectively.

[0069] Those skilled in the art will appreciate that the closer an order volume segment is to the peak season center segment, the closer the order volume within that segment is to the peak season center segment. The closer an order volume segment is to the off-season center segment, the closer the order volume within that segment is to the off-season center segment. Therefore, the degree of proximity between an order volume segment and the peak season center segment can be calculated using the average order volume within that order volume segment and the average order volume within the peak season center segment. The degree of proximity between an order volume segment and the off-season center segment can be calculated using the average order volume within that order volume segment and the average order volume within the off-season center segment.

[0070] Specifically, the degree of proximity k1 between the order volume segment n and the peak season center segment a is:

[0071]

[0072] The degree of proximity k2 between the order volume segment n and the off-season center segment b is:

[0073]

[0074] Where, mean(n) is the mean order volume of order volume segment n, mean(a) is the mean order volume of the peak season center segment a closest to order volume segment n, and mean(b) is the mean order volume of the off-season center segment b closest to order volume segment n. The peak season center segment a and the off-season center segment b are located on the left and right sides of order volume segment n, respectively.

[0075] In some alternative embodiments, the proximity k1 between the order volume segment n and the peak season center segment a is:

[0076] The degree of proximity k2 between the order volume segment n and the off-season center segment b is:

[0077] Where t(a,b) is the duration between the peak-season center segment a and the off-season center segment b, t(a,n) is the duration between the order volume segment n and the peak-season center segment a, and t(b,n) is the duration between the order volume segment n and the off-season center segment b. The peak-season center segment a and the off-season center segment b are the peak-season center segments and off-season center segments closest to the left and right sides of the order volume segment n, respectively.

[0078] Secondly, obtain the stability of the order volume segment.

[0079] The stability level is the difference between 1 and the ratio of the duration between the order volume segment and the nearest target center segments on its left and right sides. Specifically, the stability level is the difference between 1 and the ratio of the duration between the order volume segment and the nearest peak season center segments and off-season center segments on its left and right sides.

[0080] Specifically, the stability w of the order volume segment n n for:

[0081]

[0082] Where t(n) is the duration of order volume segment n, t(a,b) is the duration between peak-season center segment a and off-season center segment b, and peak-season center segment a and off-season center segment b are the peak-season center segments and off-season center segments closest to the left and right sides of order volume segment n, respectively.

[0083] The closer the duration ratio between the order volume segment and its nearest peak-season and off-season segments to its left and right is to 1, the longer the order volume within that segment has remained stable. This indicates that the segment is neither peak-season nor off-season. Therefore, the stability level is determined by the difference between 1 and the duration ratio between the order volume segment and its nearest peak-season and off-season segments to its left and right.

[0084] Finally, we get the peak season and off-season combined coefficients. Specifically:

[0085] The peak season consolidation coefficient h for order volume segment n n1 for:

[0086] h n1 =k1×w n ;

[0087] The off-season consolidation coefficient h for order volume segment n n2 for:

[0088] h n2 =k2×w n ;

[0089] Where k1 is the degree of proximity between the order volume segment n and the peak season center segment a, k2 is the degree of proximity between the order volume segment n and the off-season center segment b, and w n is the stability of the order volume segment n.

[0090] The present invention merges the order volume segments into the corresponding peak season center segments and off-season center segments according to the peak season merging coefficient and the off-season merging coefficient of the order volume segments, thereby obtaining the peak season segment set and the off-season segment set, making the division of the peak season and the off-season more accurate and reasonable, and avoiding the problem that when using ordered sample clustering to segment the historical order volume, the transition period between the peak season and the off-season is segmented too much and too finely, resulting in the segmentation of the period that should belong to the peak season or the off-season, thereby shortening the duration of the peak season and the off-season, and causing supply shortage or oversupply when using this as a reference for stocking.

[0091] S4. Obtain the sub-segments of each order volume segment in the peak season segment set, the off-season segment set, and the non-segment set that are most correlated with the order volume of the current time period, and configure the supply quantity based on the historical order volume after the period of the corresponding sub-segment when the judgment coefficient is greater than a preset threshold.

[0092] The preset threshold value ranges from 0.6 to 0.8, preferably 0.7. Of course, the value of the preset threshold value can be adjusted as needed.

[0093] Although steps S1 to S3 yield multiple complete peak, off-season, and flat season segments, in the actual sales process, the order volume during the current period may not be completely consistent with the order volume during each of these periods, and it is uncertain which peak, off-season, or flat season segment the current period falls into. Therefore, it is necessary to obtain a subsegment within the peak season segment set, a subsegment within the off-season segment set, and a subsegment within the flat season segment set that are most correlated with the current period's order volume. This yields three subsegments, each representing the trend most similar to the current period's order volume. Then, based on the judgment coefficient, the subsegment most similar to the current period's order volume is determined.

[0094] The specific process is as follows:

[0095] S41. Obtain three sub-segments with the highest correlation with the order volume of the current period. The three sub-segments include one sub-segment in the peak season segment set, one sub-segment in the off-season segment set, and one sub-segment in the normal season segment set.

[0096] The current period refers to the period before today. The length of the current period can be 5 days, 10 days, 15 days, etc., and can be selected according to needs.

[0097] The above sub-segments are obtained by moving the sliding window in each order volume segment in the peak season segment set, the off-season segment set and the non-segment set, where the size of the sliding window is equal to the duration of the current period.

[0098] The degree of correlation is the Pearson correlation coefficient obtained by calculating the order volume of the current period and the order volume of the sub-segments of each order volume segment in the peak season segment set, the off-season segment set, and the non-segment set (i.e., the off-season segment set).

[0099] The Pearson correlation coefficient measures the correlation between two variables, with values ​​ranging from -1 to 1. A value of 1 indicates a perfect positive correlation between the two variables, a value of -1 indicates a perfect negative correlation, and a value of 0 indicates no linear correlation between the two variables. Using the Pearson correlation coefficient to calculate correlation is simple, efficient, and widely applicable.

[0100] Specifically, the calculation formula of the Pearson correlation coefficient r is as follows:

[0101]

[0102] Where x i y is any order quantity of a sub-segment of each order quantity segment in the peak season segment set, off-season segment set, and non-segment set, i is the volume of any order in the current period; The mean order volume of the sub-segments of each order volume segment in the peak season segment set, off-season segment set, and non-segment set; is the average order volume of the current period; n is the duration of the current period.

[0103] Through the above calculation formula, we can get multiple Pearson correlation coefficients r1 corresponding to the peak season segment set, multiple Pearson correlation coefficients r2 corresponding to the off-season segment set, and multiple Pearson correlation coefficients r3 corresponding to the flat season segment set. Take the maximum value of r1, r2, and r3, that is, the value closest to 1, and get r 1max , r 2max , r 3max Then r 1max The corresponding sub-segment is the sub-segment of the peak season segment set that has the highest correlation with the order volume of the current period, r 2max The corresponding sub-segment is the sub-segment of the off-season segment set that has the highest correlation with the order volume of the current period, r 3max The corresponding sub-segment is the sub-segment of the shoulder season segment set that has the highest correlation with the order volume of the current period.

[0104] The three sub-segments obtained at this time only represent the trend of changes in the order volume that is closest to the current period, and do not represent the actual order volume. That is, the order volume within this sub-segment may be significantly different from the order volume of the current period. Therefore, it is necessary to determine the sub-segment that is closest to the order volume of the current period.

[0105] Of course, other methods can also be used to calculate the degree of correlation, such as the Spearman correlation coefficient, the Kendall correlation coefficient, etc.

[0106] S42. Allocate supply quantity based on the historical order quantity after the period of the corresponding sub-segment where the judgment coefficient is greater than the preset threshold.

[0107] The corresponding sub-segment obtained in this step is the sub-segment closest to the current time period. It can be considered that the order volumes of the two will be similar in the next period of time. Therefore, the supply volume can be configured with reference to the historical order volume after the period in which the sub-segment is located.

[0108] The judgment coefficient is the product of the ratio of the current period duration to the order volume period duration of the corresponding sub-segment and the degree of similarity. The degree of similarity indicates the similarity between the order volume of the current period and the order volume within the sub-segment.

[0109] The calculation process of the judgment coefficient is as follows:

[0110] First, get the proximity.

[0111] The degree of closeness c between the order volume of the current period and the order volume in the sub-segment is:

[0112]

[0113] Where, mean(Q i ) is the current period Q i The mean order quantity, mean(Q j ) is the sub-segment Q j The average order volume of the sub-segment Q j With the current period Q i of equal length.

[0114] In this step, by comparing the similarity between the average order volume of the current period and the average order volume of the sub-segment, the degree of closeness between the order volume in the current order volume segment and the order volume of the sub-segment can be reflected.

[0115] In some alternative embodiments, the degree of closeness may also be represented by a normalized value of the difference between the order volume of the current period and the sum of the order volumes of the sub-segments, wherein the sum may also be the median or the mode.

[0116] Next, obtain the ratio of the duration of the current period to the duration of the order volume segment in the corresponding sub-segment. The ratio d is:

[0117] Where t(i) is the duration of the current period, and t(j) is the duration of the order volume segment in which the corresponding sub-segment falls. The longer the current period, the larger the ratio d, and the closer the current period is to the order volume segment in which the corresponding sub-segment falls. This prevents misjudgments caused by excessively high judgment coefficients due to a short timeframe.

[0118] Finally, the judgment coefficient is calculated.

[0119] The judgment coefficient p is: p = c × d;

[0120] Where c is the closeness between the order volume of the current period and the order volume in the sub-segment, and d is the ratio of the duration of the current period to the duration of the order volume segment in the corresponding sub-segment.

[0121] In some embodiments, configuring the supply quantity based on the historical order quantity after the period of the corresponding sub-segment where the judgment coefficient is greater than a preset threshold includes:

[0122] First, the corresponding historical order volumes after the period of the sub-segment corresponding to the current period in different regions are obtained respectively, and the corresponding historical order volumes are used as the required supply volumes for the corresponding regions.

[0123] Specifically, by performing steps S1 to S4 above on the historical order volume for the same product in different regions, the sub-segments of each order volume segment corresponding to the peak season, off-season, and non-segment sets for the current time period in each region are obtained. The corresponding historical order volume after the time period in which the sub-segment falls is used as the required supply quantity for that region. This is because different regions have regional differences, and different sub-segments corresponding to the current time period will also result in different required supply quantities. Therefore, when configuring the required supply quantity for different regions, the priority of the configuration also varies.

[0124] Secondly, sort the required supply quantities from large to small, and allocate the required supply quantities for the corresponding regions in sequence according to the order of the required supply quantities.

[0125] This step prioritizes supply to regions with the highest volume. Once supply is complete for that region, it then moves on to other regions, ultimately completing supply to all regions. This allows for a rational allocation of required supply across regions, prioritizing supply to regions with the highest volume, and ultimately achieving maximum economic benefits.

[0126] In the digital supply chain management method of the present invention, order volume segments are merged into corresponding peak season central segments and off-season central segments according to the merging coefficient of the peak season and the merging coefficient of the off-season, thereby obtaining a peak season segment set, an off-season segment set and a flat season segment set, making the division of the peak season, off-season and flat season more accurate and reasonable, and according to the degree of correlation and the judgment coefficient, a sub-segment of the peak season segment set, off-season segment set or flat season segment set that is most similar to the order volume of the current period can be obtained, and the supply volume is configured according to the historical order volume after the period where the corresponding sub-segment is located, so that reasonable stocking can be achieved, avoiding the problem of supply shortage or oversupply caused by the inability to effectively distinguish the transition segment between the peak season and the off-season in the existing technology, and stocking with the peak season and the off-season as a reference.

[0127] like Figure 2 As shown, according to the second aspect of the present invention, a digital supply chain management system is also provided. The system includes a memory and a processor. The processor executes the computer program stored in the memory to implement the digital supply chain management method described in the first aspect of the present invention.

[0128] The system further includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and thus will not be described in detail here.

[0129] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented by computer-readable / executable instructions stored or otherwise retained by such a computer-readable medium.

[0130] In the description of this specification, “a plurality of” means at least two, for example, two, three or more, etc., unless otherwise clearly defined.

[0131] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A digital supply chain management method, characterized in that: The steps include: Get the historical order volume of goods; Use the ordered sample clustering method to cluster the historical order volume and obtain multiple order volume segments; Obtain peak season center segments and off-season center segments from multiple order volume segments, calculate the peak season merging coefficient and off-season merging coefficient of any segment in the remaining order volume segments, and merge any segment with the peak season center segment or the off-season center segment to obtain a peak season segment set and an off-season segment set; the remaining unmerged order volume segments are a non-segment set; wherein the merging coefficient is the product of the degree of proximity and stability between the order volume segment and the target center segment; the stability is the difference between 1 and the ratio of the duration between the order volume segment and the nearest target center segments on its left and right sides; the target center segment is the peak season center segment or the off-season center segment; The sub-segments of each order volume segment in the peak season segment set, off-season segment set and non-segment set that have the highest degree of correlation with the order volume of the current time period are obtained respectively, and the supply quantity is configured according to the historical order volume after the time period of the corresponding sub-segment when the judgment coefficient is greater than the preset threshold; wherein the judgment coefficient is the product of the ratio of the current time period length and the order volume segment length of the corresponding sub-segment and the degree of proximity; the degree of proximity represents the similarity between the order volume of the current time period and the order volume in the sub-segment.

2. The digital supply chain management method according to claim 1, characterized in that: The peak season center segment is the order volume segment corresponding to the maximum value of the order volume mean among all order volume segments, and the off-season center segment is the order volume segment corresponding to the minimum value of the order volume mean among all order volume segments.

3. The digital supply chain management method according to claim 1, characterized in that: The calculation of the peak season merging coefficient and the off-season merging coefficient of any segment in the remaining order volume segments to merge any segment with the peak season center segment or the off-season center segment includes: When the peak season merging coefficient of any of the remaining order volume segments is greater than the preset merging coefficient, the order volume segment is merged into the peak season center segment; When the off-season merging coefficient of any segment in the remaining order volume segments is greater than the preset merging coefficient, the order volume segment is merged into the off-season central segment.

4. The digital supply chain management method according to claim 1, characterized in that: The degree of proximity k1 between the order volume segment n and the peak season center segment a is: The degree of proximity k2 between the order volume segment n and the off-season center segment b is: Where, mean(n) is the mean order volume of order volume segment n, mean(a) is the mean order volume of the peak season center segment a closest to order volume segment n, and mean(b) is the mean order volume of the off-season center segment b closest to order volume segment n. The peak season center segment a and the off-season center segment b are located on the left and right sides of order volume segment n, respectively.

5. The digital supply chain management method according to claim 1, characterized in that: The degree of correlation is the Pearson correlation coefficient obtained by respectively calculating the order volume of the current period and the order volume of the sub-segments of each order volume segment in the peak season segment set, the off-season segment set and the non-segment set.

6. The digital supply chain management method according to claim 1, characterized in that: The sub-segments are obtained by moving a sliding window on each order volume segment in the peak season segment set, the off-season segment set, and the non-segment set; wherein the size of the sliding window is equal to the duration of the current period.

7. The digital supply chain management method according to claim 1, characterized in that: The degree of closeness c is: Where, mean(Q i ) is the current period Q i The mean of the order quantity, mean(Q j ) is the sub-segment Q j The mean of the order quantity, and the sub-segment Q j With the current period Q i of equal length.

8. The digital supply chain management method according to claim 1, characterized in that: The configuration of the supply quantity based on the historical order quantity after the period of the corresponding sub-segment where the judgment coefficient is greater than the preset threshold includes: Obtain the corresponding historical order volume after the period of the corresponding sub-segment of the current period in different regions, and use the corresponding historical order volume as the required supply volume for the corresponding region; Sort the required supply quantities from largest to smallest, and allocate the required supply quantities for the corresponding regions in the order of the required supply quantities.

9. The digital supply chain management method according to claim 1, characterized in that: The historical order volume is the historical order volume in days.

10. A digital supply chain management system, comprising a memory and a processor, characterized in that: The processor executes the computer program stored in the memory to implement the digitalization-based supply chain management method according to any one of claims 1 to 9.

Images