Cigarette visiting weekly sales plan decomposition algorithm for sales terminal customers

Abstract

The invention discloses a cigarette visiting weekly sales plan decomposition algorithm for sales terminal clients. The algorithm comprises the following steps: analyzing client data, and counting the number of corresponding clients under client characteristics; inputting the decomposition parameters and the attributes of the to-be-decomposed product and regulation into a plan decomposition algorithm, and obtaining the putting amount corresponding to each gear in the sales week of product and regulation decomposition; inputting the decomposed data into a logistics sorting multiple verification algorithm, and verifying and adjusting the total putting amount of the specifications at each gear, so that the putting amount of each specification at each gear meets a corresponding multiple relationship; and inputting the multiple-adjusted decomposed data and the efficiency requirement into a logistics sorting efficiency verification algorithm, and verifying and adjusting the specification delivery amount, so that the total specification delivery amount meets the efficiency requirement. Through the algorithm, the enterprise can more scientifically decompose the whole sales target to each terminal client step by step, thereby ensuring the accurate implementation of the sales task, and improving the efficiency and precision of sales execution.

Description

Technical Field

[0001] The present invention relates to the technical field of sales plan decomposition algorithms, and particularly to a cigarette visit sales weekly sales plan decomposition algorithm for sales terminal customers. Background Art

[0002] In the cigarette industry, the formulation and execution of sales plans are key links in enterprise operations. Cigarette sales not only require accurate market forecasts but also need to flexibly respond to changing market demands and consumer behaviors. To ensure the achievement of sales targets, cigarette enterprises usually formulate annual or quarterly sales targets and further break them down into monthly or weekly sales plans. These sales plans need to be accurately decomposed to each sales terminal customer to achieve specific sales tasks. However, existing sales plan decomposition methods often rely on manual statistics and empirical judgments, lacking a scientific and systematic decomposition mechanism, resulting in waste of resources, deviation in execution, and difficulties in achieving targets.

[0003] With the continuous progress of information technology and data analysis technology, data-driven sales plan decomposition has gradually become a trend in the industry's development. Through in-depth analysis of multi-dimensional information such as historical sales data, market trends, customer behaviors, and external environments, sales plans can be decomposed more accurately, enabling scientific forecasts and customized sales strategies for each sales terminal customer. However, most current technologies are limited to static analysis models and lack the ability of dynamic adjustment and real-time optimization. In addition, traditional decomposition methods have not effectively considered the differences among sales terminals and the particularities of regional markets, resulting in low efficiency and target deviation during the execution of sales plans. Summary of the Invention

[0004] The purpose of the present invention is to: in view of the above existing problems, provide a cigarette visit sales weekly sales plan decomposition algorithm for sales terminal customers, which combines modern data analysis technology with industry requirements, aiming to achieve refined decomposition of sales plans through precise processing of multi-dimensional data and dynamically adjust the delivery strategy.

[0005] The technical solution of the present invention is as follows:

[0006] A cigarette visit sales weekly sales plan decomposition algorithm for sales terminal customers includes the following steps:

[0007] Analyze customer data and count the number of customers corresponding to the customer characteristics.

[0008] Input the decomposition parameters and the product specifications to be decomposed into the plan decomposition algorithm to obtain the sales volume of the product specifications decomposed into the visit week.

[0009] Input the decomposed data into the logistics sorting multiple verification algorithm, verify and adjust the total delivery volume of each product specification at each gear so that the total sales volume of each product specification at each gear meets the corresponding multiple relationship;

[0010] Input the decomposed data after adjusting the multiple and the efficiency requirements into the logistics sorting efficiency verification algorithm, verify and adjust the proportion of the total sales volume of the product specification so that the total sales volume of the product specification meets the efficiency requirements.

[0011] Further, the planned decomposition algorithm includes:

[0012] Judge the stop-bidding flag of the product specification. If the stop-bidding flag is 1, execute the decomposition algorithm for this product specification. If the stop-bidding flag is 0, ignore the decomposition of this product specification in the current visit and sales week;

[0013] Judge whether the source utilization rate in the product specification parameters exists. If it does not exist, calculate the source utilization rate R of the product specification: R = S / P, where S is the sales and order volume of the product specification corresponding to the previous week, and P is the delivery volume of the product specification corresponding to the previous week;

[0014] Calculate the planned delivery volume PN of the product specification in the current visit and sales week: PN = L·s·125 / R, where s is the market status coefficient corresponding to the product specification, and s includes: "popular": 1.0, "tight": 0.9, "average": 0.8, "loose": 0.7, "weak": 0.6; L is the remaining delivery volume of the product specification corresponding to this month;

[0015] Determine the actual delivery volume of the product specification through the delivery strategy. If the strategy is 1, the actual delivery volume p of the product specification is the effective inventory I corresponding to the product specification. If the strategy is 0, the actual delivery volume p of the product specification is the minimum value between the effective inventory I and the planned delivery volume PN;

[0016] Query the delivery weights cW of each urban network group gear and the delivery weights rW of each rural network group gear; Since cW and rW are arrays with a length of 30, if there are consecutive 0s in the array, repair the holes where cW and rW weights are continuously 0 through the hole repair algorithm;

[0017] Count the number of effective customers participating in the delivery of each product specification in each gear for the urban network and rural network;

[0018] Calculate the delivery volume in each gear for each product specification according to the gear weight of the actual delivery volume;

[0019] After delivering all the product specifications in all gears, if there is still a remainder in the actual delivery volume, perform cyclic supplementary delivery 1 for the gears with non-zero weights until the remaining volume is less than the number of effective customers in all gears; Obtain the preliminary decomposed data of the product specification in all gears.

[0020] Further, the query of the delivery weights cW of each urban network group gear and the delivery weights rW of each rural network group gear includes:

[0021] If the decomposition parameters include the delivery weights for each gear, then determine the delivery mode. The specific steps are as follows:

[0022] If the delivery mode of this product specification is 'gear', and the delivery mode of the cigarette price category corresponding to this product specification is 'gear' or 'gear + market', then cW and rW take the weight data in the decomposition parameters. At this time, set the delivery mode pM of this product specification to 'gear';

[0023] If the delivery mode corresponding to this product specification is 'gear + urban network + rural network', and the delivery mode of the cigarette price category corresponding to this product specification is 'gear + market', then cW and rW take the weight data in the decomposition parameters. If the delivery mode of the cigarette price category corresponding to this product specification is 'gear', then cW takes the weight data in the decomposition parameters, and cW takes the weight data in the decomposition parameters multiplied by the coefficient 0.6. At this time, set the delivery mode pm of this product specification to 'gear + urban network + rural network';

[0024] If the delivery mode corresponding to this product specification is 'gear + urban network', then cW takes the weight data in the decomposition parameters, and rW takes 0. At this time, set the delivery mode pM of this product specification to 'gear + urban network';

[0025] If the delivery mode corresponding to this product specification is 'gear + rural network', then cW takes 0, and rW takes the weight data in the decomposition parameters. At this time, set the delivery mode pM of this product specification to 'gear + rural network';

[0026] If the decomposition parameters do not include the delivery weights for each gear, then automatically read the historical weekly and monthly delivery data. If there is corresponding product specification data for this product specification in the historical delivery data, then cW and rW are respectively equal to the delivery quantities of the corresponding urban network gear and rural network gear in the historical data for each historical product specification. If there is no corresponding product specification data for this product specification in the historical delivery data, then cW and rW of this product specification are respectively equal to the delivery quantities of the corresponding urban network gear and rural network gear in the historical price category data for this cigarette price category. At this time, when cW is equal to rW, then set the delivery type pM of this product specification to 'gear', otherwise, set pM to 'gear + market'.

[0027] Further, the hole repair algorithm infers and fills in the missing data by analyzing the values of adjacent elements, including:

[0028] Forward inference: Starting from the beginning of the array, check the relationship between the current element and the previous element; if the previous element is positive and the current element is empty, then fill it with the value of the previous element;

[0029] Backward repair: If the subsequent element is positive, use this positive value to fill the current or the next hole;

[0030] Mean filling: In the case of multiple consecutive holes, use the mean of adjacent positive values for filling;

[0031] Iterative processing: Repeat this process until all holes are reasonably filled.

[0032] Furthermore, the statistics of the number of effective customers participating in the urban and rural network placements for each product specification in each gear include:

[0033] If pM is "gear + market", determine whether there are outflowing districts or counties for this product specification. If there are outflowing districts or counties, the number of effective customers cC participating in the urban network placement is equal to the number of customers c0 in the urban network corresponding to each gear in the customer data B1 of the market type minus the number of customers c in the outflowing districts or counties corresponding to the urban network in each gear in the administrative division customer data B3. Similarly, the number of effective customers rC participating in the rural network placement can be calculated;

[0034] If there are no outflowing districts or counties, the number of effective customers cC participating in the urban network placement is equal to the number of customers c0 in the urban network corresponding to each gear segment in the data B1. Similarly, the number of effective customers rC participating in the rural network placement can be calculated;

[0035] If pM is "gear", determine whether there are outflowing districts or counties for this product specification. If there are outflowing districts or counties, the number of effective customers C participating in the placement is equal to the total number of customers c0 corresponding to each gear and each market type in the data B1 minus the total number of customers c1 in all outflowing districts or counties corresponding to each market type in each gear segment in the data B3; if there are no outflowing districts or counties, the number of effective customers C participating in the placement is equal to the total number of customers c0 corresponding to each gear and each market type in the data B1.

[0036] Furthermore, the calculation of the placement volume in each gear according to the gear weight for the actual placement volume of each product specification includes:

[0037] If the product specification placement mode pM is "gear + market", calculate the placement volume pi of the product specification in each gear:

[0038]

[0039]

[0040] Among them, p is the actual placement volume corresponding to this product specification, pi is the placement volume array, mT is the corresponding market type, and its element i needs to be rounded to an integer. When 0 < i < 1, i takes 1;

[0041] If the placement mode pM of the product specification is "gear", calculate the placement volume pi of the product specification in each gear:

[0042]

[0043] Among them, cW is equal to rW.

[0044] Furthermore, the logistics sorting multiple verification algorithm includes:

[0045] Query the sorting type of the product specification data. One sorting type lb corresponds to the same sorting multiple T. Product specifications with lb being 1 do not need to perform multiple verification. Filter out the list of product specifications pS belonging to the same lb. Among the decomposition data of pS, further filter out the set of product specifications ps with the delivery quantity greater than 0 at this gear g;

[0046] Calculate the comprehensive weighted score mS of the market status:

[0047] mS = 0.6·R + 0.4·s

[0048] Among them, R is the source utilization rate of this product specification, and s is the market status coefficient;

[0049] Calculate the total number of delivery items tS corresponding to this gear g for all product specifications in ps; if the total number of delivery items is greater than or equal to T, then calculate the remainder rst = tS % T. If the remainder rst is less than or equal to the upper limit L, it conforms to the rule and no verification is required. If the remainder rst is greater than the upper limit L, and the verification intensity strength is'strong','middle' or 'weak', then perform the increment +1 operation T - rst times on the delivery quantity of the gear g corresponding to the product specifications in ps. If the increment is unsuccessful, and the verification intensity is strong or middle, then perform the decrement -1 operation rst - L times; if the total number of delivery items is less than T, if the verification intensity strength is'strong','middle' or 'weak', first perform the increment +1 operation T - tS times. If the increment is unsuccessful, and the verification intensity strength is'strong', then perform the decrement -1 operation tS times; obtain the decomposition data after multiple verification.

[0050] Furthermore, the increment +1 operation includes:

[0051] Sort the weighted scores mS of the product specifications from large to small;

[0052] Count the number of customers s0 of each market type corresponding to this gear in data B1. If there are outflowing districts and counties, count the number of customers s1 of the outflowing districts and counties corresponding to each market type at this gear in data B3. The effective number of customers s of this product specification at this gear is s = s0 - s1;

[0053] Take the product specifications in the sorted queue in turn. If the remaining inventory I is greater than or equal to the effective number of customers s of the customer at this gear g, then add 1 delivery item for this product specification at this gear;

[0054] Loop the previous step until the remaining inventory is insufficient or the incremental delivery times are met, then stop the loop;

[0055] The decrement - 1 operation includes:

[0056] Sort the weighted scores mS of the product specifications from small to large;

[0057] In data B1, count the number of customers s0 of each market type corresponding to this gear. If there are outflow districts and counties, in data B3, count the number of customers s1 of the outflow districts and counties corresponding to each market type under this gear. The effective number of customers s of this product specification in this gear is s = s0 - s1;

[0058] Successively take the product specifications in the sorted queue. If the remaining inventory I is greater than or equal to the effective number of customers s of the customer in this gear g, then reduce the delivery of this product specification by 1 in this gear;

[0059] Loop the previous step until the decrement delivery times are met or the delivery volume in this gear is 0, then stop the loop.

[0060] Furthermore, the logistics sorting efficiency verification algorithm includes:

[0061] Construct a dictionary of cigarette codes participating in the delivery and sorting types, obtain the logistics efficiency indicators in the decomposition parameters, and construct the logistics efficiency indicators for 5 days in the call - visit week. R0 is the proportion of horizontal machines, R1 is the proportion of horizontal machine group A, R2 is the proportion of horizontal machine group B, R3 is the proportion of vertical machines, and R4 is the proportion of vertical manual cigarettes. Among them, R0, R1, R2, R3, and R4 are all arrays with a length of 5;

[0062] In data B8_i, calculate the number of customers c0 of the corresponding market type of the customer gear on the i - th day. If there is an outflow, in data B9_i, calculate the number of customers c1 of all outflow districts and counties corresponding to each market type under this gear. The effective number of customers c of this product specification in each gear on the i - th day is c = c0 - c1. In the decomposed data after multiple - factor verification, multiply the number of delivery pieces of each product specification in each gear by the effective number of customers c of the corresponding gear and sum to obtain the total delivery volume of this product specification on the i - th day. Then sum the total delivery volumes of all product specifications to obtain the total sales volume Q;

[0063] Search for the product specifications ps1 whose sorting equipment belongs to horizontal machines 1, 2, and 3 by cigarette code, calculate the total delivery volume W1 of these product specifications on the i - th day, search for the product specifications ps2 whose sorting equipment belongs to horizontal machines 4 and 5 by cigarette code, calculate the total delivery volume W2 of these product specifications on the i - th day, search for the product specifications ps3 whose sorting equipment belongs to vertical machines by cigarette code, calculate the total delivery volume L1 of these product specifications on the i - th day, and search for the product specifications ps4 of square cigarettes by cigarette code, calculate the total delivery volume L2 of these product specifications on the i - th day;

[0064] Verify the logistics sorting efficiency within every 5 days of the verification and sales week, calculate the proportion according to the following formula. Among them, the efficiency verification conditions are: R2 <= R2_max, R3 <= R3_max, R4 <= R4_max.

[0065] R0 = (W1 + W2) / Q,

[0066] R1 = W1 / Q,

[0067] R2 = W2 / Q,

[0068] R3 = L1 / Q,

[0069] R4 = L2 / Q,

[0070] Adjust the delivery volume of product specifications that do not meet the efficiency requirements. If R4 > R4_max, reduce the delivery volume of product specifications with type number lb being 1. If R3 > R3_max, count the total delivery volume of cigarette coding types lb being 2, 42, and 43 respectively, screen out the product specification ps corresponding to the lb with the largest total delivery volume on the same day, and reduce the delivery volume of ps. If R2 > R2_max, count the total delivery volume of cigarette coding types lb being 41 and 44 respectively, screen out the product specification ps corresponding to the lb with the largest total delivery volume on the same day, and reduce the delivery volume of ps. Loop through this step until it is the same as the previous adjustment result or meets the verification conditions, then no further adjustment is required.

[0071] Among them, the algorithm principle for reducing the delivery volume is as follows:

[0072] According to the product specification ps for which the delivery volume is to be reduced, query the product specifications ps in the decomposition data after multiple - factor verification of each product specification in the list where the delivery volume of this product specification at this gear g is greater than 0.

[0073] Calculate the weighted score mS according to the source utilization rate R and market status coefficient s of the product specifications in ps.

[0074] Calculate the total number of delivery lines tS corresponding to this gear g for all product specifications in ps; if the total number of delivery lines is greater than or equal to T, calculate the remainder rst = tS % T. If the remainder rst is greater than 0, and the verification intensity strong is'strong' or'middle', then perform a single - reduction operation of - 1 on the delivery volume of the product specifications in ps corresponding to this gear g. If the remainder rst is less than 0, and the verification intensity strong is'strong', then perform T single - reduction operations of - 1 on the delivery volume of the product specifications in ps corresponding to this gear g. Do not handle other situations.

[0075] Calculate the total sorting efficiency in the decomposition data after logistics efficiency verification.

[0076] Furthermore, the analysis of customer data includes the following steps:

[0077] According to the customer grades and market types, count the number of customers in each market type corresponding to each grade, and record this data as B1;

[0078] According to the customer grades, market types, and order cycles, count the number of customers in each order cycle corresponding to each market type in all grades, and record this data as B2;

[0079] According to the customer grades, market types, and administrative regions, count the number of customers in each administrative region corresponding to each market type in all grades, and record this data as B3;

[0080] According to the customer grades, market types, administrative regions, and order cycles, count the number of customers in each administrative region corresponding to each order cycle under each market type in all grades, and record this data as B4;

[0081] Count the input customer order cycles, and determine whether each order cycle exists in the "order cycle" of the customer data. If not, delete the corresponding customer order cycle, and record this data as B5;

[0082] According to the obtained customer order cycles, count the number of customers in each order cycle corresponding to each market type in all grades in data B2, and record this data as B6;

[0083] According to the obtained customer order cycles, count the number of customers in each order cycle corresponding to each administrative region under each market type in all grade segments in data B4, and record this data as B7;

[0084] According to the obtained customer order cycles of each day, count the number of customers in the i-th day order cycle corresponding to each market type in all grades in data B2, and record this data as B8_i;

[0085] According to the obtained customer order cycles of each day, count the number of customers in the i-th day order cycle corresponding to each administrative region under each market type in all grades in data B4, and record this data as B9_i.

[0086] The beneficial effects of the present invention compared with the existing technology are:

[0087] A cigarette visit and sales weekly sales plan decomposition algorithm for sales terminal customers can accurately decompose the sales target into a specific weekly sales plan through a data-driven algorithm model. Under the condition of meeting the logistics sorting multiple and logistics sorting efficiency, it ensures that the sales plan is flexible, accurate, and scientific, reducing human deviation. Compared with the traditional sales plan decomposition method, this application can optimize resource allocation, improve sales efficiency, and provide a rapid market response ability according to historical actual sales data, customer needs, and market changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 This is the overall flowchart of the algorithm of this application.

[0089] Figure 2 This is the flowchart of the plan decomposition algorithm of this application.

[0090] Figure 3 This is the flowchart of the logistics sorting multiple verification algorithm of this application.

[0091] Figure 4 This is the flowchart of the logistics efficiency verification algorithm of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0092] It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0093] The features and performance of the present invention will be further described in detail below in conjunction with the embodiments.

[0094] Please refer to Figures 1-4 , a cigarette visit and sales weekly sales plan decomposition algorithm for sales terminal customers, as Figure 1 shown, includes the following steps:

[0095] Analyze customer data and count the number of customers corresponding to the customer characteristics; the customer characteristics include: "customer grade", "market type", "administrative division", "order cycle";

[0096] Analyzing customer data includes the following steps:

[0097] According to the customer grades and market types, count the number of customers in each market type corresponding to each grade, and record this data as B1;

[0098] According to the customer grades, market types, and order cycles, count the number of customers in each order cycle corresponding to each market type in all grades, and record this data as B2;

[0099] According to the customer grades, market types, and administrative regions, count the number of customers in each administrative region corresponding to each market type in all grades, and record this data as B3;

[0100] According to the customer grades, market types, administrative regions, and order cycles, count the number of customers in each order cycle corresponding to each administrative region under each market type in all grades, and record this data as B4;

[0101] Count the input customer order cycles, and determine whether each order cycle exists in the "order cycle" of the customer data. If not, delete the corresponding customer order cycle, and record this data as B5.

[0102] Based on the obtained customer order cycles, count the number of customers in each order cycle corresponding to each market type in all grades in data B2, and record this data as B6;

[0103] Based on the obtained customer order cycles, count the number of customers in each order cycle corresponding to each administrative region under each market type in all grade segments in data B4, and record this data as B7;

[0104] Based on the obtained daily customer order cycles, count the number of customers in the i-th day order cycle corresponding to each market type in all grades in data B2, and record this data as B8_i;

[0105] Based on the obtained daily customer order cycles, count the number of customers in the i-th day order cycle corresponding to each administrative region under each market type in all grades in data B4, and record this data as B9_i.

[0106] Input the decomposition parameters and the product specifications to be decomposed into the planned decomposition algorithm to obtain the sales volume of the product specifications decomposed to the visit and sales week; the decomposition parameters include: visit and sales week code, number of visit and sales weeks in the month to which the visit and sales week belongs, delivery types corresponding to cigarettes at each price level, weight of urban network and rural network group grades, and attributes of the product specifications to be decomposed. Among them, cigarettes at each price level include: high-price cigarettes, high-end cigarettes, first to fifth category cigarettes, and the attributes of the product specifications to be decomposed include: product specification code, corresponding last week's delivery volume P, corresponding last week's sales and order volume S, corresponding remaining delivery volume L this month, effective inventory I corresponding to this round of delivery, cigarette price level, decomposition type, decomposition unit, sorting multiple, market status, outflow districts and counties, stop delivery flag, delivery type, and source utilization rate R.

[0107] As Figure 2 shown, the planned decomposition algorithm includes:

[0108] Judge the stop delivery flag of the product specification. If the stop delivery flag is 1, execute the decomposition algorithm for this product specification. If the stop delivery flag is 0, ignore the decomposition of this product specification in this visit and sales week;

[0109] Judge whether the source utilization rate in the product specification parameters exists. If it does not exist, calculate the source utilization rate R of the product specification: R = S / P, where S is the corresponding last week's sales and order volume of the product specification, and P is the corresponding last week's delivery volume of the product specification;

[0110] Calculate the planned delivery volume PN of the product specification in this visit and sales week: PN = L·s·125 / R,

[0111] where s is the market status coefficient corresponding to the product specification, and s includes: "popular": 1.0, "tight": 0.9, "stable": 0.8, "loose": 0.7, "weak": 0.6; L is the corresponding remaining delivery volume of the product specification this month; product specifications with low source utilization rate reduce the delivery volume, and product specifications with high source utilization rate increase the delivery volume. The compensation coefficient 1.25 ensures that product specifications with a source utilization rate higher than 80% can obtain sufficient delivery.

[0112] Determine the actual delivery volume of the product specification through the delivery strategy. If the strategy is 1, the actual delivery volume p of the product specification is the corresponding effective inventory. If the strategy is 0, the actual delivery volume p of the product specification is the minimum value between the effective inventory I and the planned delivery volume PN;

[0113] Query the delivery weights cW of each urban network group grade and the delivery weights rW of each rural network group grade; including:

[0114] If the decomposition parameters include the delivery weights of each grade, judge the delivery mode. The specific steps are as follows:

[0115] If the placement mode of this product specification is 'grade', and the placement mode of the cigarette price category corresponding to this product specification is 'grade' or 'grade + market', then cW and rW take the weight data in the decomposition parameters. At this time, set the placement mode pM of this product specification to 'grade';

[0116] If the placement mode corresponding to this product specification is 'grade + market (urban network + rural network)', and the placement mode of the cigarette price category corresponding to this product specification is 'grade + market', then cW and rW take the weight data in the decomposition parameters. If the placement mode of the cigarette price category corresponding to this product specification is 'grade', then cW takes the weight data in the decomposition parameters, and cW takes the weight data in the decomposition parameters multiplied by the coefficient 0.6. At this time, set the placement mode pM of this product specification to 'grade + market (urban network + rural network)';

[0117] If the placement mode corresponding to this product specification is 'grade + market (urban network)', then cW takes the weight data in the decomposition parameters, rW takes 0. At this time, set the placement mode pM of this product specification to 'grade + market (urban network)';

[0118] If the placement mode corresponding to this product specification is 'grade + market (rural network)', then cW takes 0, rW takes the weight data in the decomposition parameters. At this time, set the placement mode pM of this product specification to 'grade + market (rural network)'.

[0119] If the placement weights of each grade are not included in the decomposition parameters, automatically read the historical weekly and monthly placement data. If there is corresponding product specification data for this product specification in the historical placement data, then cW and rW are respectively equal to the placement quantities of the corresponding urban network grade and rural network grade in the historical product specifications in the historical data. If there is no corresponding product specification data for this product specification in the historical placement data, then cW and rW of this product specification are respectively equal to the placement quantities of the corresponding urban network grade and rural network grade in the historical price category data of this cigarette price category. At this time, when cW is equal to rW, set the placement type pM of this product specification to 'grade', otherwise, set pM to 'grade + market'.

[0120] Since cW and rW are arrays of length 30 and there are consecutive zeros in the arrays, repair the holes with zero weights in cW and rW through the hole repair algorithm; the hole repair algorithm infers and fills in the missing data by analyzing the values of adjacent elements, including:

[0121] Forward inference: Start from the beginning of the array and check the relationship between the current element and the previous element. If the previous element is positive and the current element is empty, fill it with the value of the previous element;

[0122] Backward repair: If the subsequent element is positive, use this positive value to fill the current or the next hole;

[0123] Mean filling: In the case of multiple consecutive holes, use the mean of adjacent positive values for filling;

[0124] Iterative processing: Repeat this process until all holes are reasonably filled.

[0125] Statistically count the number of effective customers participating in the urban and rural network placements for each product specification at each gear; If pM is "gear + market", determine whether there are outflowing districts or counties for this product specification. If there are outflowing districts or counties, the number of effective customers cC participating in the urban network placement is equal to the number of customers c0 in the urban network corresponding to each gear in the customer data B1 of the market type minus the number of customers c in the outflowing districts or counties corresponding to the urban network under each gear in the administrative division customer data B3. Similarly, the number of effective customers rC participating in the rural network placement can be calculated;

[0126] If there are no outflowing districts or counties, the number of effective customers cC participating in the urban network placement is equal to the number of customers c0 in the urban network corresponding to each gear segment in the data B1 in the table. Similarly, the number of effective customers rC participating in the rural network placement can be calculated;

[0127] If pM is "gear", determine whether there are outflowing districts or counties for this product specification. If there are outflowing districts or counties, the number of effective customers C participating in the placement is equal to the total number of customers c0 corresponding to each gear and each market type in the data B1 minus the total number of customers c1 in all outflowing districts or counties corresponding to each market type under each gear segment in the data B3; If there are no outflowing districts or counties, the number of effective customers C participating in the placement is equal to the total number of customers c0 corresponding to each gear and each market type in the data B1.

[0128] Calculate the placement quantity in each gear for the actual placement quantity of each product specification according to the gear weight; including:

[0129] If the product specification placement mode pM is "gear + market", calculate the placement quantity pi of the product specification in each gear:

[0130]

[0131] Among them, p is the actual placement quantity corresponding to this product specification, pi is the placement quantity array, mT is the corresponding market type, and its element i needs to be rounded to an integer. When 0 < i < 1, i takes 1;

[0132] If the placement mode pM of the product specification is "gear", calculate the placement quantity pi of the product specification in each gear:

[0133]

[0134] Among them, cW is equal to rW.

[0135] After the product specifications are input for all gears, if there is still a surplus in the actual input quantity, then perform cyclic supplementary input 1 for the gears with non-zero weights until the remaining quantity is less than the number of customers in all gears; obtain the preliminary decomposition data of the product specifications for all gears.

[0136] Input the decomposition data into the logistics sorting multiple verification algorithm, verify and adjust the total input quantity of the product specifications for each gear, so that the total sales volume of each product specification for each gear meets the corresponding multiple relationship;

[0137] As Figure 3 shown, the logistics sorting multiple verification algorithm includes:

[0138] Query the sorting type of the product specification data. One sorting type lb corresponds to the same sorting multiple T. The product specifications with lb being 1 do not need to perform multiple verification. Screen out the list of product specifications pS belonging to the same lb. In the decomposition data of pS, further screen out the set of product specifications ps with the input quantity greater than 0 for this gear g;

[0139] Calculate the comprehensive weighted score mS of the market status:

[0140] mS = 0.6·R + 0.4·s

[0141] where R is the source utilization rate of this product specification, and s is the market status coefficient;

[0142] Calculate the total number of input items tS corresponding to all product specifications in ps for this gear g; if the total number of input items is greater than or equal to T, then calculate the remainder rst = tS % T. If the remainder rst is less than or equal to the upper limit L, it meets the rule and does not need verification. If the remainder rst is greater than the upper limit L, and the verification intensity strength is'strong','middle' or 'weak', then perform the operation of incrementing +1 by T - rst times on the input quantity of the gear g corresponding to the product specifications in ps. If the increment is unsuccessful, and the verification intensity is strong or middle, then perform the operation of decrementing -1 by rst - L times. If the total number of input items is less than T, if the verification intensity strength is'strong','middle' or 'weak', first perform the operation of incrementing +1 by T - tS times. If the increment is unsuccessful, and the verification intensity strength is'strong', then perform the operation of decrementing -1 by tS times. Obtain the decomposition data after multiple verification.

[0143] The operation of incrementing +1 includes:

[0144] Sort the weighted scores mS of the product specifications from large to small;

[0145] Count the number of customers s0 of each market type corresponding to this gear in Data B1. If there are outflow districts and counties, count the number of customers s1 of the outflow districts and counties corresponding to each market type at this gear in Data B3. The effective number of customers s of this product specification at this gear is s = s0 - s1;

[0146] Take the product specifications in the sorting queue in turn. If the remaining inventory I is greater than or equal to the effective number of customers s of the customer at this gear g, then add 1 delivery of this product specification at this gear;

[0147] Loop the previous step until the remaining inventory is insufficient or the incremental delivery times are met, then stop the loop;

[0148] The decrement - 1 operation includes:

[0149] Sort the weighted scores mS of the product specifications from small to large;

[0150] Count the number of customers s0 of each market type corresponding to this gear in Data B1. If there are outflow districts and counties, count the number of customers s1 of the outflow districts and counties corresponding to each market type at this gear in Data B3. The effective number of customers s of this product specification at this gear is s = s0 - s1;

[0151] Take the product specifications in the sorting queue in turn. If the remaining inventory I is greater than or equal to the effective number of customers s of the customer at this gear g, then reduce 1 delivery of this product specification at this gear;

[0152] Loop the previous step until the decrement delivery times are met or the delivery quantity at this gear is 0, then stop the loop.

[0153] Input the decomposed data after adjusting the multiple and the efficiency requirements into the logistics sorting efficiency verification algorithm, verify and adjust the proportion of the total sales volume of the product specifications so that the total sales volume of the product specifications meets the efficiency requirements.

[0154] As Figure 4 shown, the logistics sorting efficiency verification algorithm includes:

[0155] Construct a dictionary of cigarette codes participating in the delivery and sorting types, obtain the logistics efficiency indicators in the decomposition parameters, and construct the logistics efficiency indicators for 5 days in the call week. R0 is the proportion of the horizontal machine, R1 is the proportion of horizontal machine group A, R2 is the proportion of horizontal machine group B, R3 is the proportion of the vertical machine, and R4 is the proportion of the vertical manual cigarettes. Among them, R0, R1, R2, R3, and R4 are all arrays with a length of 5;

[0156] Calculate the quantity c0 of the market type corresponding to the customer grade on the i-th day in data B8_i. If there is an outflow, calculate the quantity c1 of customers in each market type corresponding to all outflow districts and counties under this grade in data B9_i. The effective customer quantity c of each grade of this product specification on the i-th day is c = c0 - c1. In the decomposed data after multiple-check verification, multiply the number of delivery pieces of each product specification in each grade by the effective customer quantity c of the corresponding grade and sum them to obtain the total delivery volume of this product specification on the i-th day. Then sum the total delivery volumes of each product specification to obtain the total sales volume Q;

[0157] Search for product specifications ps1 whose sorting equipment belongs to horizontal machines 1, 2, and 3 (i.e., the sorting codes are '51' and '53') according to the cigarette code, and calculate the total delivery volume (to customers) W1 of these product specifications on the i-th day. Search for product specifications ps2 whose sorting equipment belongs to horizontal machines 4 and 5 (i.e., the sorting codes are '41' and '44') according to the cigarette code, and calculate the total delivery volume (to customers) W2 of these product specifications on the i-th day. Search for product specifications ps3 whose sorting equipment belongs to vertical machines (i.e., the sorting codes are '1', '2', '42', and '43') according to the cigarette code, and calculate the total delivery volume (to customers) L1 of these product specifications on the i-th day. Search for product specifications ps4 of block cigarettes (i.e., the sorting code is '1') according to the cigarette code, and calculate the total delivery volume (to customers) L2 of these product specifications on the i-th day;

[0158] Verify the logistics sorting efficiency every 5 days in the visit and sales week, and calculate the proportion according to the following formula. Among them, the efficiency verification conditions are: R2 <= R2_m□□, R3 <= R3_max, R4 <= R4_max,

[0159] R0 = (W1 + W2) / Q,

[0160] R1 = W1 / Q,

[0161] R2 = W2 / Q,

[0162] R3 = L1 / Q,

[0163] R4 = L2 / Q,

[0164] Adjust the delivery volume of product specifications that do not meet the efficiency requirements. If R4 > R4_max, reduce the delivery volume of product specifications with type number lb being 1. If R3 > R3_max, count the total delivery volume of product specifications with cigarette code types lb being 2, 42, and 43 respectively, select the product specification ps corresponding to the lb with the largest total delivery volume on the same day, and reduce the delivery volume of ps. If R2 > R2_max, count the total delivery volume of product specifications with cigarette code types lb being 41 and 44 respectively, select the product specification ps corresponding to the lb with the largest total delivery volume on the same day, and reduce the delivery volume of ps. Loop through this step until it is the same as the previous adjustment result or meets the verification conditions, then no subsequent adjustment is required;

[0165] Among them, the algorithm principle for reducing the delivery quantity (reducing the remainder of the multiple relationship) is as follows:

[0166] According to the product specification ps of the delivery quantity to be reduced input, query the product specifications ps in the decomposition data after multiple verification where the delivery quantity of each product specification in the list at this gear g is greater than 0;

[0167] Calculate the weighted score mS based on the supply source utilization rate R and the market status coefficient s of the product specifications in ps;

[0168] Calculate the total number of delivery items tS corresponding to this gear g for all product specifications in ps; if the total number of delivery items is greater than or equal to T, calculate the remainder rst = tS % T. If the remainder rst is greater than 0, and the verification intensity strong is'strong' or'middle', then perform a single reduction -1 operation on the delivery quantity of the product specifications in ps corresponding to this gear g; if the remainder rst is less than 0, and the verification intensity strong is'strong', then perform T reduction -1 operations on the delivery quantity of the product specifications in ps corresponding to this gear g; do not process in other cases;

[0169] In the decomposition data after logistics efficiency verification, calculate the total sorting efficiency.

[0170] The above embodiments only represent the specific implementation manners of the present application, and the description is relatively specific and detailed, but it should not be construed as a limitation on the protection scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the technical solution of the present application, several deformations and improvements can still be made, and these all belong to the protection scope of the present application.

Claims

1. A weekly sales plan decomposition algorithm for cigarette sales to sales terminal customers, characterized in that: The following steps are involved: Analyze customer data and count the number of customers corresponding to each customer characteristic; Input the decomposition parameters and the product specifications to be decomposed into the plan decomposition algorithm to obtain the sales volume of the product specifications decomposed into the sales visit week; Input the decomposed data into the logistics sorting multiple verification algorithm to verify and adjust the total delivery volume of each product specification at each level so that the total sales volume of each product specification at each level meets the corresponding multiple relationship; The decomposed data and efficiency requirements after adjustment are input into the logistics sorting efficiency verification algorithm to verify and adjust the proportion of the total sales volume of the product specification so that the total sales volume of the product specification meets the efficiency requirements.

2. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 1 is characterized in that: The plan decomposition algorithm includes: Determine the stop flag of the product specification. If the stop flag is 1, execute the decomposition algorithm for the product specification. If the stop flag is 0, ignore the decomposition of the product specification in this sales visit week. Determine whether the supply utilization rate in the product specification parameters exists. If not, calculate the supply utilization rate R of the product specification: R = S / P, where S is the sales order quantity of the product specification corresponding to last week, and P is the supply quantity of the product specification corresponding to last week; Calculate the planned release volume PN of the product specification in this sales week: PN = L·s·125 / R, where s is the market status coefficient corresponding to the product specification, including: "pretty": 1.0, "tight": 0.9, "flat": 0.8, "loose": 0.7, "soft": 0.6; L is the remaining release volume of the product specification this month; The actual release quantity of the product specification is determined by the release strategy. If the strategy is 1, the actual release quantity p of the product specification is the effective inventory I corresponding to the product specification. If the strategy is 0, the actual release quantity p of the product specification is the minimum value between the effective inventory I and the planned release quantity PN. Query the delivery weight cW of each urban network group and the delivery weight rW of each rural network group; since cW and rW are arrays of length 30, if there are consecutive 0s in the array, the hole patching algorithm is used to patch the holes where the cW and rW weights are consecutively 0; Count the number of effective customers participating in the launch of each product specification in urban and rural networks at each level; The actual release quantity of each product specification is calculated according to the weight of each gear. After all gears have been allocated for the product specification, if there is still a remaining actual amount of investment, the gears with weights other than 0 will be cyclically supplemented with an investment of 1 until the remaining amount is less than the number of valid customers in all gears; thus, the preliminary decomposition data of the product specification in all gears is obtained.

3. The decomposition algorithm of the weekly cigarette sales plan for sales terminal customers according to claim 2 is characterized in that: The querying of the delivery weight cW of each urban network group gear and the delivery weight rW of each rural network group gear includes: If the decomposition parameters include the delivery weights of each gear, the delivery mode is determined. The specific steps are as follows: If the product specification's delivery mode is 'gear', and the delivery mode of the cigarette price category corresponding to the product specification is 'gear' or 'gear + market', cW and rW take the weight data in the decomposition parameters, and the delivery mode pM of the product specification is set to 'gear'; If the delivery mode corresponding to the product specification is 'gear + urban network + rural network', and the delivery mode of the cigarette price category corresponding to the product specification is 'gear + market', cW and rW take the weight data in the decomposition parameters, if the delivery mode of the cigarette price category corresponding to the product specification is 'gear', cW takes the weight data in the decomposition parameters, cW takes the weight data in the decomposition parameters multiplied by the coefficient 0.6, at this time, the delivery mode pM of the product specification is set to 'gear + urban network + rural network'; If the delivery mode corresponding to the product specification is "gear + city network", cW takes the weight data in the decomposition parameter, rW takes 0, and the delivery mode pW of the product specification is set to "gear + city network"; If the delivery mode corresponding to the product specification is "gear + rural network", cW takes 0, rW takes the weight data in the decomposition parameter, and the delivery mode PM of the product specification is set to "gear + rural network"; If the decomposition parameters do not include the delivery weights of each gear, the historical weekly and monthly delivery data will be automatically read. If the product specification has corresponding product specification data in the historical delivery data, cW and rW will be equal to the delivery volume of each historical product specification corresponding to the urban network gear and rural network gear in the historical data respectively. If the product specification does not have corresponding product specification data in the historical delivery data, cW and rW of the product specification will be equal to the delivery volume of the urban network gear and rural network gear corresponding to the cigarette price category in the historical price category data respectively. At this time, when cW is equal to rW, the delivery type pM of the product specification is set to 'gear', otherwise, pM is set to 'gear+market'.

4. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 2 or 3, characterized in that: The hole-filling algorithm infers and fills in missing data by analyzing the values ​​of neighboring elements, including: Forward extrapolation: Start at the beginning of the array and check the relationship between the current element and the previous element; if the previous element is positive and the current element is empty, fill it with the value of the previous element; Backward patching: If the subsequent element is positive, use the positive value to fill the current or next hole; Mean filling: In the case of multiple consecutive holes, the mean of adjacent positive values ​​is used to fill them; Iterative processing: Repeat this process until all holes are reasonably filled.

5. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 4 is characterized in that: The statistics of the number of effective customers participating in the urban and rural network launches of each product specification in each tier include: If pM is "Level + Market", determine whether there are outflow districts and counties for this product specification. If there are outflow districts and counties, the number of effective customers participating in the launch in the urban network cC is equal to the number of customers c0 of the urban network corresponding to each level in the market type customer data B1 minus the number of customers c of the urban network corresponding to the outflow districts and counties under each level in the administrative division customer data B3. Similarly, the number of effective customers rC participating in the launch in the rural network can be calculated; If there is no outflow district or county, the number of effective customers participating in the urban network cC is equal to the number of customers c0 of the urban network corresponding to each gear segment in data B1. Similarly, the number of effective customers participating in the rural network rC can be calculated; If pM is "gear", determine whether there are outflow counties for this product specification. If there are outflow counties, the number of effective customers C participating in the placement is equal to the total number of customers c0 corresponding to each gear and each market type in data B1 minus the total number of customers c1 in all outflow counties corresponding to each market type under each gear segment in data B3; if there are no outflow counties, the number of effective customers C participating in the placement is equal to the total number of customers c0 corresponding to each gear and each market type in data B1.

6. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 4 is characterized in that: The calculation of the placement quantity in each gear according to the gear weight for the actual placement quantity of each product specification includes: If the product specification placement mode pM is "gear + market", calculate the placement quantity pi of the product specification in each gear: Where p is the actual placement quantity corresponding to this product specification, pi is the placement quantity array, mT is the corresponding market type, and its element i needs to be rounded to an integer. When 0 < i < 1, i is taken as 1; If the placement mode pM of the product specification is "gear", calculate the placement quantity pi of the product specification in each gear: Where cW is equal to rW.

7. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 1 is characterized in that: The logistics sorting multiple verification algorithm includes: Query the sorting type of the product specification data. One sorting type lb corresponds to the same sorting multiple T. Product specifications with lb equal to 1 do not need to perform multiple verification. Screen out the list of product specifications pS belonging to the same lb. In the decomposition data of pS, further screen out the set of product specifications ps with a placement quantity greater than 0 in this gear g; Calculate the comprehensive weighted score mS of the market status: mS = 0.6·R + 0.4·s Where R is the source utilization rate of this product specification, and s is the market status coefficient; Calculate the total number of placement lines tS corresponding to this gear g for all product specifications in ps; if the total number of placement lines is greater than or equal to T, then calculate the remainder rst = tS % T. If the remainder rst is less than or equal to the upper limit L, it conforms to the rule and does not need verification. If the remainder rst is greater than the upper limit L, and the verification intensity strength is'strong','middle' or 'weak', then perform a T - rst times increment +1 operation on the placement quantity of the gear g corresponding to the product specifications in ps. If the increment is unsuccessful, and the verification intensity is strong or middle, then perform a rst - L times decrement -1 operation; if the total number of placement lines is less than T, if the verification intensity strength is'strong','middle' or 'weak', first perform a T - tS times increment +1 operation. If the increment is unsuccessful, and the verification intensity strength is'strong', then perform a tS times decrement -1 operation; obtain the decomposition data after multiple verification.

8. The cigarette sales weekly plan decomposition algorithm for sales terminal customers according to claim 7, characterized in that: The increment +1 operation includes: Sort the weighted scores mS of the product specifications from large to small; Statistically count the number of customers s0 corresponding to each market type of this gear in data B1. If there are outflow counties, statistically count the number of customers s1 in the outflow counties corresponding to each market type under this gear in data B3. The number of effective customers of this product specification in this gear is s = s0 - s1; Successively take the product specifications in the sorted queue. If the remaining inventory I is greater than or equal to the number of effective customers s of the customer in this gear g, then this product specification increases 1 placement line in this gear. Repeat the previous step until the remaining inventory is insufficient or the number of incremental delivery times is met, then stop the loop; The decrement -1 operation includes: Sort the weighted scores mS of the product specifications from small to large; In data B1, the number of customers s0 corresponding to each market type in this gear is counted. If there are outflow districts and counties, the number of customers s1 corresponding to each market type in this gear is counted in data B3. The number of effective customers of this product specification in this gear is s=s0-s1; Take the product specifications in the sorting queue in turn. If the remaining inventory I is greater than or equal to the number of valid customers s at the customer's gear position g, then reduce the number of products in this gear by 1; The previous step is repeated until the number of reduction times is met or the amount of delivery at this gear is 0, then the cycle stops.

9. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 1, characterized in that: The logistics sorting efficiency verification algorithm includes: Construct a dictionary of cigarette codes and sorting types involved in the launch, obtain the logistics efficiency index in the decomposition parameters, and construct the logistics efficiency index for the five days of the sales week. R0 is the proportion of horizontal machines, R1 is the proportion of horizontal group A, R2 is the proportion of horizontal group B, R3 is the proportion of vertical machines, and R4 is the proportion of vertical handmade cigarettes. R0, R1, R2, R3, and R4 are all arrays of length 5. In data B8_i, calculate the number c0 of market types corresponding to the customer level on the i-th day. If there is outflow, calculate the number c1 of customers corresponding to all outflow districts and counties for each market type under the level in data B9_i. The number of effective customers of each level for the product specification on the i-th day c=c0-c1. In the decomposed data after multiple verification, multiply the number of items placed in each level for each product specification by the number of effective customers c of the corresponding level and sum them up to get the total amount of the product specification placed on the i-th day. Then sum the total amount of the product specifications placed to get the total sales volume Q. According to the cigarette code, find the specification ps1 of the sorting equipment belonging to horizontal machines 1, 2, and 3, and calculate the total amount of these specifications put in on the i-th day W1; according to the cigarette code, find the specification ps2 of the sorting equipment belonging to horizontal machines 4 and 5, and calculate the total amount of these specifications put in on the i-th day W2; according to the cigarette code, find the specification ps3 of the sorting equipment belonging to the vertical machine, and calculate the total amount of these specifications put in on the i-th day L1; according to the cigarette code, find the specification ps4 of the square cigarette, and calculate the total amount of these specifications put in on the i-th day L2; Verify the logistics sorting efficiency every 5 days of the sales visit week and calculate the percentage according to the following formula, where the efficiency verification conditions are: R2 <= R2_max, R3 <= R3_max, R4 <= R4_max, R0=(W1+W2) / Q, R1=W1 / Q, R2=W2 / Q, R3=L1 / Q, R4=L2 / Q, Adjust the quantity of the specifications that do not meet the efficiency requirements. If R4>R4_max, reduce the quantity of the specifications with type number lb as 1. If R3>R3_max, count the total quantity of cigarettes with code types lb of 2, 42, and 43, filter out the specifications ps corresponding to the lb with the largest total quantity on that day, and reduce the quantity of ps. If R2>R2_max, count the total quantity of cigarettes with code types lb of 41 and 44, filter out the specifications ps corresponding to the lb with the largest total quantity on that day, and reduce the quantity of ps. Repeat this step until the result is the same as the previous round of adjustment or the verification conditions are met, and no subsequent adjustment is required. Among them, the algorithm principle for reducing the amount of delivery is: According to the input product specification ps whose release quantity is to be reduced, query the product specifications ps whose release quantity at the gear position g is greater than 0 in the decomposed data of each product specification in the list after multiple verification; Calculate the weighted score mS based on the supply utilization rate R and market status coefficient s of the product specifications in ps; Calculate the total number of items tS for the gear g of all product specifications in ps; if the total number of items is greater than or equal to T, calculate the remainder rst = tS%T. If the remainder rst is greater than 0 and the verification strength strong is 'strong' or 'middle', then the amount of items for the gear g corresponding to the product specifications in ps is reduced by 1 times; if the remainder rst is less than 0 and the verification strength strong is 'strong', then the amount of items for the gear g corresponding to the product specifications in ps is reduced by T times; no processing is performed in other cases; Calculate the total sorting efficiency from the decomposed data after the logistics efficiency verification.

10. The cigarette sales weekly sales plan decomposition algorithm for sales terminal customers according to claim 1, characterized in that: Analyzing customer data includes the following steps: According to customer levels and market types, count the number of customers of all levels corresponding to each market type, and record this data as B1; According to customer tiers, market types and order cycles, count the number of customers in each order cycle corresponding to each market type in all tiers, and record this data as B2; According to customer tiers, market types and administrative divisions, count the number of customers in each administrative division corresponding to each market type in all tiers, and record this data as B3; According to customer level, market type, administrative division and order cycle, count the number of customers in each order cycle in each administrative division under each market type in all levels, and record this data in B4; Count and input customer order cycles, and determine whether each order cycle exists in the "Order Cycle" in the customer data. If not, delete the corresponding customer order cycle and record this data in B5. According to the obtained customer order cycle, the number of customers in each market type in all gears corresponding to the customer order cycle is counted in data B2, and this data is recorded as B6; According to the obtained customer order cycle, the number of customers in each administrative region corresponding to the customer order cycle in each market type in all gear segments is counted in data B4, and this data is recorded as B7; According to the obtained customer order cycle every day, the number of customers in each market type in all gears corresponding to the order cycle on the i-th day is counted in the data B2, and this data is recorded as B8_i; Based on the obtained daily customer ordering cycle, the number of customers in the ordering cycle on the i-th day corresponding to each administrative division under each market type in all gears is counted in data B4, and this data is recorded as B9_i.

Images