CN114066491A - Method and device for determining goods replenishment strategy based on multi-cycle total sales forecast - Google Patents
Method and device for determining goods replenishment strategy based on multi-cycle total sales forecast Download PDFInfo
- Publication number
- CN114066491A CN114066491A CN202010742505.0A CN202010742505A CN114066491A CN 114066491 A CN114066491 A CN 114066491A CN 202010742505 A CN202010742505 A CN 202010742505A CN 114066491 A CN114066491 A CN 114066491A
- Authority
- CN
- China
- Prior art keywords
- sales
- probability distribution
- determining
- distribution data
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Cash Registers Or Receiving Machines (AREA)
Abstract
The application relates to a method and a device for determining an article replenishment strategy based on multi-cycle total sales forecast. The method comprises the following steps: acquiring historical sales data of the sales of the articles; inputting historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantile points in the quantile prediction model; determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data; determining second cumulative probability distribution data of the total goods forecast sales of all periods with set quantity according to the first probability distribution data of each period based on a calculation method of joint probability distribution; determining a predicted value of the total commodity sales amount in each set number period according to the second cumulative probability distribution data; and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles. By adopting the method, the accuracy of the supply chain article replenishment data can be improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a computer device, and a storage medium for determining an item restocking policy based on a multi-cycle predicted total sales.
Background
The goods sales forecasting is an important technology in the field of supply chains, is widely applied to various aspects of replenishment, inventory management, logistics management and cost control in the field of supply chains, particularly quantile forecasting of goods sales, can give a variation range and occurrence probability of goods demand, and is particularly useful for inventory optimization. The supply chain field usually considers extrapolating the quantile forecasting results of a plurality of cycle sales sums in a plurality of cycles (such as days, weeks and months) in the future, and the habit of forecasting the goods sales is carried out by taking one cycle as a unit, but the sum of the quantiles of the plurality of cycle sales sums is not equal to the quantile of the plurality of cycle sales sums, so an effective method for forecasting the quantiles of the plurality of cycle sales sums is needed.
However, mean value prediction is adopted for predicting the commodity sales volume sums in multiple cycles at present, namely, the mean values of the sums in multiple cycles are obtained according to the mean value prediction sums in multiple cycles, and the quantile prediction result cannot be obtained; or the commodity sales amount is assumed to obey normal distribution, quantiles of the multiple cycles and quantiles are calculated according to the additivity of the normal distribution, but the commodity sales amount in an actual scene probably does not obey the normal distribution, so that the total commodity sales amount prediction accuracy of the multiple cycles is low, and the supply chain commodity replenishment data is inaccurate.
Disclosure of Invention
In view of the above, it is necessary to provide a prediction method, apparatus, computer device and storage medium for determining an item restocking strategy based on a multi-cycle predicted total sales amount, which can improve accuracy of supply chain item restocking data.
A method of determining an item restocking strategy based on a predicted total sales volume for a plurality of cycles, the method comprising:
acquiring historical sales data of the sales of the articles;
inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantiles in the quantile prediction model;
determining first probability distribution data of the article sales in the corresponding period according to the first cumulative probability distribution data;
determining second cumulative probability distribution data of the total goods forecast sales of all the periods with set number according to the first probability distribution data of each period based on a calculation method of joint probability distribution;
determining a predicted value of the total predicted commodity sales amount of each set number period according to the second cumulative probability distribution data;
and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
In one embodiment, the inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the article in each period of a set number of periods according to the number of quantiles in the quantile prediction model includes:
inputting the historical sales data into a quantile prediction model, and determining the cumulative probability value of the predicted value of the item sales corresponding to each quantile point in each period in a set number period;
and determining first cumulative probability distribution data of the item sales volume in each period in a set number period according to the predicted value and the cumulative probability value of the item sales volume corresponding to each branch point.
In one embodiment, the determining, from each of the first cumulative probability distribution data, first probability distribution data corresponding to the sales of the item over the period includes:
and determining a prediction probability value corresponding to the predicted value of the commodity sales volume according to the cumulative probability value of each quantile point based on the conversion relation between the probability distribution and the cumulative probability distribution to obtain first probability distribution data of the commodity sales volume in each period.
In one embodiment, the joint probability distribution-based calculation method determines, from the first probability distribution data for each cycle, second cumulative probability distribution data of predicted total sales of the items for all the set number of cycles, including:
determining second probability distribution data of the total sales of the articles in all the set number periods according to the first probability distribution data based on a calculation method of joint probability distribution;
and determining second cumulative probability distribution data of the corresponding total commodity sales according to the second probability distribution data of the total commodity sales of each set number period based on the conversion relation between the probability distribution and the cumulative probability distribution.
In one embodiment, the method for calculating a joint probability distribution based on determining a second probability distribution data of total sales of the items for all the set number of cycles according to each of the first probability distribution data includes:
based on a calculation method of joint probability distribution, carrying out weighted calculation on the prediction probability value of the corresponding item sales volume in each first probability distribution data to obtain the probability value of the total sales volume of all the items in a set number period;
and aggregating the probability values with the same total commodity sales in the set number periods to obtain second probability distribution data of the total commodity sales in all the set number periods.
In one embodiment, the method further comprises:
sampling the predicted value of the commodity sales volume in each period in a set number period for preset times to obtain a sampled commodity sales volume set;
simulating the sampling article sales volume set to obtain a simulation result set;
and determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile.
In one embodiment, the determining replenishment strategy data of the corresponding item according to the predicted value of the total sales volume of the item includes:
acquiring the current inventory allowance of the article;
and updating the current inventory allowance according to the predicted value of the total item sales in each set quantity period.
An apparatus for determining an item restocking strategy based on a predicted total sales volume for a plurality of cycles, the apparatus comprising:
the acquisition module is used for acquiring historical sales volume data of the sales volume of the article;
the prediction module is used for inputting the historical sales data into a quantile prediction model and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to quantile parameter values of the quantile prediction model;
a first determining module, configured to determine, according to each of the first cumulative probability distribution data, first probability distribution data corresponding to an amount of sales of the item in the period;
a second determining module, configured to determine, according to the first probability distribution data of each period, second cumulative probability distribution data of the total predicted sales of the items for all the periods of the set number based on a calculation method of a joint probability distribution;
a third determining module, configured to determine a predicted value of the total commodity sales in each of the set number periods according to the second cumulative probability distribution data;
and the fourth determination module is used for determining replenishment strategy data of the corresponding article according to the predicted total sales value of the article.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
acquiring historical sales data of the sales of the articles;
inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantiles in the quantile prediction model;
determining first probability distribution data of the article sales in the corresponding period according to the first cumulative probability distribution data;
determining second cumulative probability distribution data of the total goods forecast sales of all the periods with set number according to the first probability distribution data of each period based on a calculation method of joint probability distribution;
determining a predicted value of the total predicted commodity sales amount of each set number period according to the second cumulative probability distribution data;
and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring historical sales data of the sales of the articles;
inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantiles in the quantile prediction model;
determining first probability distribution data of the article sales in the corresponding period according to the first cumulative probability distribution data;
determining second cumulative probability distribution data of the total goods forecast sales of all the periods with set number according to the first probability distribution data of each period based on a calculation method of joint probability distribution;
determining a predicted value of the total predicted commodity sales amount of each set number period according to the second cumulative probability distribution data;
and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
The method, the device, the computer equipment and the storage medium for determining the goods replenishment strategy based on the multi-cycle predicted total sales are provided; obtaining historical sales data of goods sales; inputting historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantile points in the quantile prediction model; determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data; determining second cumulative probability distribution data of the total goods forecast sales of all periods with set quantity according to the first probability distribution data of each period based on a calculation method of joint probability distribution; determining a predicted value of the total commodity sales amount in each set number period according to the second cumulative probability distribution data; determining replenishment strategy data of corresponding articles according to the predicted value of the total sales volume of the articles; the total goods sales amount of multiple cycles is determined according to the quantile result, the total goods sales amount prediction accuracy of the multiple cycles is improved, and the accuracy of the supply chain goods replenishment data is improved.
Drawings
FIG. 1 is a flow diagram illustrating a method for determining an item restocking strategy based on a predicted total sales volume for multiple cycles, according to one embodiment;
FIG. 2 is a schematic flow diagram of the quantile prediction model predicting quantile points in one embodiment;
FIG. 3 is a flow chart illustrating a method for determining an item restocking strategy based on a multi-cycle forecasted total sales volume in another embodiment;
FIG. 4 is a block diagram of an embodiment of an apparatus for determining an item restocking strategy based on a multi-cycle forecasted total sales;
FIG. 5 is a block diagram of an apparatus for determining an item restocking strategy based on a multi-cycle predicted total sales volume in another embodiment;
FIG. 6 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In one embodiment, as shown in fig. 1, a method for determining an item restocking policy based on a multi-cycle predicted total sales volume is provided, and this embodiment is illustrated by applying the method to a terminal, it is to be understood that the method may also be applied to a server, and may also be applied to a system including the terminal and the server, and is implemented by interaction between the terminal and the server. In this embodiment, the method includes the steps of:
The historical sales data are data with time granularity attributes, and the time granularity comprises days, weeks, months and the like; for example, the historical sales data for acquiring the sales of the article may be historical sales data with a time granularity of days, or historical sales data with a time granularity of weeks.
And 104, inputting the historical sales data into the quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in the set number period according to the number of the quantile points in the quantile prediction model.
Wherein the quantile prediction model is used to predict the item sales value of the item at a given N quantile points in each of H cycles (e.g., days, weeks, or months) in the future; for example, the forecast values of the item sales of 5 quantiles, 25 quantiles, 50 quantiles, 75 quantiles and 95 quantiles in a future cycle are 10, 20, 45, 80 and 98 pieces according to the quantile forecast model and the historical sales data; wherein, the value of each quantile in the N quantiles is 1/N, 2/N. The value of N is adjusted according to the required precision of cumulative probability distribution; for example, if N is 4, then there are 25, 50, 75, 100 quantile points; if N is 10, then there are 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 quantile points, and the cumulative probability value distribution data of the commodity sales in a single period can be determined according to the N quantile points. Cumulative probability distribution probability data FX(x) Expressed as P (X is less than or equal to X), and represents that the predicted value X of the commodity sales is not largeProbability at x. For example, 4 quantiles are obtained, the value of the 3 rd quantile is 3/4, namely 75%, and the corresponding predicted commodity sales value is 55, so that the probability that the commodity sales predicted on the 3 rd day is not greater than 55 is 75%.
Specifically, before historical sales data are input into a quantile prediction model, determining time granularity, a set quantity period H and the number N of quantiles of the goods sales data to be predicted according to the time granularity of the historical sales data; and inputting the historical sales data into the quantile prediction model to obtain the predicted value of the item sales of each quantile in N quantiles in each period in H periods, and determining first cumulative probability distribution data of the item sales in each period in the H periods according to the value of each quantile and the predicted value of the corresponding item sales. For example, as shown in fig. 2, in an embodiment, a schematic diagram of predicting an article sales prediction value of a quantile through a quantile prediction model is shown, the obtained historical sales data of the article sales are obtained by taking days as time granularity, the historical sales data are input into the quantile prediction model, the quantile prediction is performed on the articles for H days in the future, the prediction values of N different quantiles are obtained, H × N prediction values are obtained, according to the N quantiles of each day, first cumulative probability distribution data of the day can be determined, and a total of H days is obtained, that is, first cumulative probability distribution data of H groups is obtained.
And 106, determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data.
Specifically, based on the conversion relationship between the probability distribution and the cumulative probability distribution, the prediction probability value of the corresponding item sales volume prediction value is determined according to the cumulative probability value of each quantile point, so as to obtain first probability distribution data of the item sales volume in each period, namely N quantiles given by a quantile prediction model according to the first cumulative probability distribution data,
represents the cumulative probability of the ith quantile point correspondence,
represents the h (h) th future<H) The ith quantile point value of the demand forecast of each period, namely the commodity sales forecast value,
indicating that the sales of the articles fall in the interval
Based on the conversion relation between the probability distribution and the cumulative probability distribution, the prediction probability value of the corresponding item sales volume prediction value is determined according to the cumulative probability value of each branch point, namely the current time
When the temperature of the water is higher than the set temperature,
i.e. to indicate that the article has been sold by
Is obtained by
To obtain an article sales of
Has a probability of
Namely, the first probability distribution data of the commodity sales in the period is determined, and the first probability distribution data of the commodity sales in the future H periods can be obtained in the same way. For example, 4 quantiles are given by the quantile prediction model, the commodity sales amount corresponding to the cumulative probability of 25% is 1, the commodity sales amount corresponding to the cumulative probability of 50% is 2, the commodity sales amount corresponding to the cumulative probability of 75% is 3, and the commodity sales amount corresponding to the cumulative probability of 100% is 4; based on the conversion relation between the probability distribution and the cumulative probability distribution, the forecast probability value of the item sales 1, 2, 3, 4 is determined to be 25%, and the first probability distribution data of the item sales in the period is obtained.
And step 108, determining second cumulative probability distribution data of the total predicted sales of the goods in all the set number of periods according to the first probability distribution data of each period based on the calculation method of the joint probability distribution.
Specifically, based on a calculation method of joint probability distribution, second probability distribution data of total commodity sales of all the set number periods are determined according to the first probability distribution data; weighting and calculating the prediction probability values of the corresponding article sales volume in each first probability distribution data based on a joint probability distribution calculation method to obtain the probability values of the total sales volumes of all articles in a set number period; and aggregating the probability values with the same total commodity sales in the set number periods to obtain second probability distribution data of the total commodity sales in all the set number periods.
Optionally, second probability distribution data of the total sales of the goods for all the set number H (1 ≦ H) cycles is determined according to the acquired first probability distribution data of the sales of the goods for each of the H cycles based on a joint probability distribution calculation method. To be provided with
Represents a 1 st cycle count of
According to a probability of
Future cycle 1 sales of
Cycle number 2 of
The h cycle sales of
The probability of total sales of the article in the future h period can be obtained as follows:
in this manner, as H increases from 1 to H, the probability of various combinations of sales for the future H periodic sales is calculated, resulting in second probability distribution data for the total sales for 1-H days. And determining second cumulative probability distribution data of the corresponding total commodity sales according to the second probability distribution data of the total commodity sales of each set number period based on the conversion relation between the probability distribution and the cumulative probability distribution.
For example, it is determined from the first probability distribution data that the probability of sales of the item for 50% is 1 and the probability of sales for 50% is 2, and assuming that the probability of sales of the item for H days is 1 for the probability of sales of the item for 50% and 2 for 50%. When h is 1, the probability distribution representing the sum of sales for 1 day is: p (1) ═ 50%, the probability that the item sales amount is 1 is 50%; p (2) ═ 50%; when h is 2, the probability that the item sales amount is 2 is 50%; the probability distribution of the two-day sales sums can be determined as:
p (1, 2) = 50%. 50%, sales 1 on the first day, sales 2 on the second day;
p (2, 1) = 50%. 50%, sales on the first day 2, sales on the second day 1;
p (2, 2) = 50% ×, sales on the first day 2, sales on the second day 2;
aggregating the probability values with the same total sales of the articles in two days to obtain second probability distribution data of the total sales of the articles in two days as follows:
p (2) ═ 25%, the probability of total 2 sales on two days is 25%;
p (3) ═ 50%, the probability of total sales of 3 on two days is 50%;
p (4) ═ 25%, with a total of 4 for two days at a probability of 25%.
Based on the transition relationship between the probability distribution and the cumulative probability distribution, according to the relationship:
FX(x)＝P(X≤x)
the above relationship indicates that for each pin value, its cumulative probability equals the sum of all the probabilities not greater than its pin value, e.g., the cumulative probability of pin 3 equals the probability of pin 2 + the probability of pin 3, i.e., F (3) ═ P (X ≦ 3) ═ P (2) + P (3),
determining the corresponding second cumulative probability distribution data according to the second probability distribution data of the total sales in two days as follows:
f (2) ═ 25%, i.e., the probability of total sales not exceeding 2 on two days is 25%;
f (3) ═ 75%, the probability that the total sales per two days do not exceed 3 is 75%;
f (4) ═ 100%, and the probability that the total sales per two days do not exceed 4 is 100%.
And step 110, determining a predicted value of the total sales of the articles in each set number period according to the second cumulative probability distribution data.
Specifically, according to the second cumulative probability distribution data, the quantile result of any quantile point, that is, the predicted value of the total sales of the articles in each set number period, can be obtained. For example, a 75% quantile results in 3.
And step 112, determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
Specifically, the replenishment strategy data of the corresponding article is determined according to the predicted value and the probability of the predicted value of the total sales volume of the articles in the set number period. For example, the probability of two days in the future and the sales amount not exceeding 2 is 25%, and the probability of the total sales amount not exceeding 3 is 75%; the probability that the total sales amount does not exceed 4 is 100%, and the replenishment quantity of the total sales amount in two days is determined according to the numerical value of the probability value and the current inventory allowance of the article.
In the method for determining the goods replenishment strategy based on the multi-cycle total forecast sales, historical sales data of goods sales are obtained; inputting historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantile points in the quantile prediction model; parameters and periods of the quantile prediction model do not need to be adjusted, and distribution of goods sales data does not need to be considered; determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data; determining second cumulative probability distribution data of the total goods forecast sales of all periods with set quantity according to the first probability distribution data of each period based on a calculation method of joint probability distribution; determining a predicted value of the total commodity sales amount in each set number period according to the second cumulative probability distribution data; determining replenishment strategy data of corresponding articles according to the predicted value of the total sales volume of the articles; the total goods sales amount of multiple cycles is determined according to the quantile result, the total goods sales amount prediction accuracy of the multiple cycles is improved, and the accuracy of the supply chain goods replenishment data is improved.
In another embodiment, as shown in fig. 3, a method for determining an item restocking policy based on a multi-cycle predicted total sales volume is provided, and this embodiment is illustrated by applying the method to a terminal, it is to be understood that the method may also be applied to a server, and may also be applied to a system including the terminal and the server, and is implemented by interaction between the terminal and the server. In this embodiment, the method includes the steps of:
Step 304, inputting the historical sales data into the quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in the set number period according to the number of the quantile points in the quantile prediction model.
And step 306, determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data.
Specifically, based on the conversion relation between the probability distribution and the cumulative probability distribution, the prediction probability value corresponding to the predicted value of the commodity sales volume is determined according to the cumulative probability value of each quantile point, and first probability distribution data of the commodity sales volume in each period is obtained.
And 308, sampling the predicted value of the item sales in each period in the set number of periods for preset times to obtain a sampled item sales set.
And 310, simulating the sampling article sales volume set to obtain a simulation result set.
Specifically, according to the first probability distribution data of the commodity sales amount of each period, for example, the probability distribution area of the commodity sales amount of each period in h periods, the commodity sales amount of each period is randomly sampled according to the probability, the sampling is performed once every period, and the obtained h sampling results are the sampling commodity sales amount set. Performing weighted calculation on the h sampling results to obtain a summation result, and inputting the summation result into a simulator to obtain a primary simulation result; and performing M times of simulation on the addition result through a simulator to obtain a simulation result set. Wherein, M may be, but is not limited to, 500 times.
And step 312, determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile.
Specifically, the simulation result sets are sorted according to the sequence of numerical values from small to large, the empirical quantiles corresponding to the simulation result sets are determined in an empirical statistical mode, and the predicted value of the total commodity sales in each set quantity period is determined according to the empirical quantiles. Optionally, quantiles in the H and fractional prediction models can be adjusted, i.e., the total sales and empirical quantiles for [1, H ] cycles can be obtained.
In step 314, the current inventory balance of the item is obtained.
And step 316, updating the current inventory allowance according to the predicted value of the total goods sales in each set quantity period.
According to the method for determining the goods replenishment strategy based on the multi-period total sales forecast, historical sales data of goods sales are obtained and input to a quantile forecasting model, and first cumulative probability distribution data of the goods sales in each period in a set number period is determined according to the number of quantiles in the quantile forecasting model; determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data; sampling the predicted value of the commodity sales volume in each period in a set number period for preset times to obtain a sampled commodity sales volume set; simulating the sampling article sales volume set to obtain a simulation result set; determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile; acquiring the current inventory allowance of the article; updating the current inventory allowance according to the predicted value of the total goods sales in each set quantity period; with the increase of the number of the prediction periods, the processing performance of the terminal can be improved, and the accuracy of the prediction value of the total commodity sales amount of the set number of periods can be improved.
It should be understood that although the various steps in the flow charts of fig. 1-3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-3 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 4, there is provided an apparatus for determining an item restocking strategy based on a predicted total sales amount for a plurality of cycles, comprising: an obtaining module 402, a predicting module 404, a first determining module 406, a second determining module 408, a third determining module 410, and a fourth determining module 412, wherein:
an obtaining module 402 is configured to obtain historical sales data of the sales of the item.
And the prediction module 404 is configured to input the historical sales data into a quantile prediction model, and determine first cumulative probability distribution data of the sales of the articles in each period in a set number of periods according to a quantile parameter value of the quantile prediction model.
A first determining module 406, configured to determine first probability distribution data corresponding to the sales of the articles in the period according to each of the first cumulative probability distribution data.
A second determining module 408, configured to determine second cumulative probability distribution data of the total predicted sales of the items for all the set number of cycles according to the first probability distribution data of each cycle based on a calculation method of a joint probability distribution.
A third determining module 410, configured to determine a predicted value of the total sales of the items in each of the set number of cycles according to the second cumulative probability distribution data.
And a fourth determining module 412, configured to determine replenishment strategy data of the corresponding item according to the item predicted total sales value.
In the article replenishment strategy device determined based on the multi-cycle total predicted sales, historical sales data of the article sales are obtained; inputting historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantile points in the quantile prediction model; parameters and periods of the quantile prediction model do not need to be adjusted, and distribution of goods sales data does not need to be considered; determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data; determining second cumulative probability distribution data of the total goods forecast sales of all periods with set quantity according to the first probability distribution data of each period based on a calculation method of joint probability distribution; determining a predicted value of the total commodity sales amount in each set number period according to the second cumulative probability distribution data; determining replenishment strategy data of corresponding articles according to the predicted value of the total sales volume of the articles; the total goods sales amount of multiple cycles is determined according to the quantile result, the total goods sales amount prediction accuracy of the multiple cycles is improved, and the accuracy of the supply chain goods replenishment data is improved.
In another embodiment, as shown in fig. 5, there is provided an apparatus for determining an item restocking strategy based on a predicted total sales amount for a plurality of cycles, comprising: the obtaining module 402, the predicting module 404, the first determining module 406, the second determining module 408, the third determining module 410, and the fourth determining module 412 further include: a weighting module 414, a sampling module 416, and a simulation module 418, wherein:
the weighting module 414 is configured to perform weighting calculation on the predicted probability values of the item sales volumes corresponding to the first probability distribution data based on a calculation method of joint probability distribution to obtain probability values of total sales volumes of all items in a set number period; and aggregating the probability values with the same total commodity sales in the set number periods to obtain second probability distribution data of the total commodity sales in all the set number periods.
And the sampling module 416 is configured to sample the predicted value of the item sales amount in each period of the set number of periods for a preset number of times to obtain a sampling item sales amount set.
And the simulation module 418 is configured to simulate the sampling commodity sales volume set to obtain a simulation result set.
In one embodiment, the prediction module 404 is further configured to input the historical sales data into a quantile prediction model, and determine an accumulated probability value of a predicted value of the item sales corresponding to each quantile in each period in a set number period; and determining first cumulative probability distribution data of the item sales volume in each period in the set number periods according to the predicted value and the cumulative probability value of the item sales volume corresponding to each sub-position point.
In one embodiment, the first determining module 406 is further configured to determine a predicted probability value corresponding to the predicted value of the item sales amount according to the cumulative probability value of each quantile point based on a conversion relationship between the probability distribution and the cumulative probability distribution, so as to obtain first probability distribution data of the item sales amount in each period.
In one embodiment, the second determining module 408 is further configured to determine second probability distribution data of the total sales of the items for all the set number of cycles according to each of the first probability distribution data based on the calculation method of the joint probability distribution; and determining second cumulative probability distribution data of the corresponding total commodity sales according to the second probability distribution data of the total commodity sales of each set number period based on the conversion relation between the probability distribution and the cumulative probability distribution.
In one embodiment, the third determining module 410 is further configured to determine an empirical quantile corresponding to the simulation result set, and determine the predicted value of the total sales of the articles for each period of the set number according to the empirical quantile.
In one embodiment, historical sales data of the sales of the articles are acquired, the historical sales data are input into a quantile prediction model, and first cumulative probability distribution data of the sales of the articles in each period in a set number of periods is determined according to the number of quantiles in the quantile prediction model; determining first probability distribution data corresponding to the sales volume of the article in the period according to the first cumulative probability distribution data; sampling the predicted value of the commodity sales volume in each period in a set number period for preset times to obtain a sampled commodity sales volume set; simulating the sampling article sales volume set to obtain a simulation result set; determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile; acquiring the current inventory allowance of the article; updating the current inventory allowance according to the predicted value of the total goods sales in each set quantity period; with the increase of the number of the prediction periods, the processing performance of the terminal can be improved, and the accuracy of the prediction value of the total commodity sales amount of the set number of periods can be improved.
The specific definition of the device for determining the article replenishment strategy based on the total predicted sales amount of the multiple cycles can be referred to the above definition of the method for determining the article replenishment strategy based on the total predicted sales amount of the multiple cycles, and is not described herein again. The modules in the above-mentioned multi-cycle-based predictive total sales determination strategy apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program when executed by a processor implements a method for determining an item restocking strategy based on a predicted total sales volume for a plurality of cycles. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 6 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring historical sales data of the sales of the articles;
inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantiles in the quantile prediction model;
determining first probability distribution data of the article sales in the corresponding period according to the first cumulative probability distribution data;
determining second cumulative probability distribution data of the total goods forecast sales of all the periods with set number according to the first probability distribution data of each period based on a calculation method of joint probability distribution;
determining a predicted value of the total predicted commodity sales amount of each set number period according to the second cumulative probability distribution data;
and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
inputting the historical sales data into a quantile prediction model, and determining the cumulative probability value of the predicted value of the item sales corresponding to each quantile point in each period in a set number period;
and determining first cumulative probability distribution data of the item sales volume in each period in a set number period according to the predicted value and the cumulative probability value of the item sales volume corresponding to each branch point.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
and determining a prediction probability value corresponding to the predicted value of the commodity sales volume according to the cumulative probability value of each quantile point based on the conversion relation between the probability distribution and the cumulative probability distribution to obtain first probability distribution data of the commodity sales volume in each period.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
determining second probability distribution data of the total sales of the articles in all the set number periods according to the first probability distribution data based on a calculation method of joint probability distribution;
and determining second cumulative probability distribution data of the corresponding total commodity sales according to the second probability distribution data of the total commodity sales of each set number period based on the conversion relation between the probability distribution and the cumulative probability distribution.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
based on a calculation method of joint probability distribution, carrying out weighted calculation on the prediction probability value of the corresponding item sales volume in each first probability distribution data to obtain the probability value of the total sales volume of all the items in a set number period;
and aggregating the probability values with the same total commodity sales in the set number periods to obtain second probability distribution data of the total commodity sales in all the set number periods.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
sampling the predicted value of the commodity sales volume in each period in a set number period for preset times to obtain a sampled commodity sales volume set;
simulating the sampling article sales volume set to obtain a simulation result set;
and determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
acquiring the current inventory allowance of the article;
and updating the current inventory allowance according to the predicted value of the total item sales in each set quantity period.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring historical sales data of the sales of the articles;
inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantiles in the quantile prediction model;
determining first probability distribution data of the article sales in the corresponding period according to the first cumulative probability distribution data;
determining second cumulative probability distribution data of the total goods forecast sales of all the periods with set number according to the first probability distribution data of each period based on a calculation method of joint probability distribution;
determining a predicted value of the total predicted commodity sales amount of each set number period according to the second cumulative probability distribution data;
and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
In one embodiment, the computer program when executed by the processor further performs the steps of:
inputting the historical sales data into a quantile prediction model, and determining the cumulative probability value of the predicted value of the item sales corresponding to each quantile point in each period in a set number period;
and determining first cumulative probability distribution data of the item sales volume in each period in a set number period according to the predicted value and the cumulative probability value of the item sales volume corresponding to each branch point.
In one embodiment, the computer program when executed by the processor further performs the steps of:
and determining a prediction probability value corresponding to the predicted value of the commodity sales volume according to the cumulative probability value of each quantile point based on the conversion relation between the probability distribution and the cumulative probability distribution to obtain first probability distribution data of the commodity sales volume in each period.
In one embodiment, the computer program when executed by the processor further performs the steps of:
determining second probability distribution data of the total sales of the articles in all the set number periods according to the first probability distribution data based on a calculation method of joint probability distribution;
and determining second cumulative probability distribution data of the corresponding total commodity sales according to the second probability distribution data of the total commodity sales of each set number period based on the conversion relation between the probability distribution and the cumulative probability distribution.
In one embodiment, the computer program when executed by the processor further performs the steps of:
based on a calculation method of joint probability distribution, carrying out weighted calculation on the prediction probability value of the corresponding item sales volume in each first probability distribution data to obtain the probability value of the total sales volume of all the items in a set number period;
and aggregating the probability values with the same total commodity sales in the set number periods to obtain second probability distribution data of the total commodity sales in all the set number periods.
In one embodiment, the computer program when executed by the processor further performs the steps of:
sampling the predicted value of the commodity sales volume in each period in a set number period for preset times to obtain a sampled commodity sales volume set;
simulating the sampling article sales volume set to obtain a simulation result set;
and determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring the current inventory allowance of the article;
and updating the current inventory allowance according to the predicted value of the total item sales in each set quantity period.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for determining an item restocking strategy based on a multi-cycle forecasted total sales, the method comprising:
acquiring historical sales data of the sales of the articles;
inputting the historical sales data into a quantile prediction model, and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to the number of quantiles in the quantile prediction model;
determining first probability distribution data of the article sales in the corresponding period according to the first cumulative probability distribution data;
determining second cumulative probability distribution data of the total goods forecast sales of all the periods with set number according to the first probability distribution data of each period based on a calculation method of joint probability distribution;
determining a predicted value of the total predicted commodity sales amount of each set number period according to the second cumulative probability distribution data;
and determining replenishment strategy data of the corresponding articles according to the predicted value of the total sales volume of the articles.
2. The method of claim 1, wherein inputting the historical sales data into a quantile predictive model, determining a first cumulative probability distribution data for the sales of the item over each of a set number of cycles based on a number of quantiles in the quantile predictive model, comprises:
inputting the historical sales data into a quantile prediction model, and determining the cumulative probability value of the predicted value of the item sales corresponding to each quantile point in each period in a set number period;
and determining first cumulative probability distribution data of the item sales volume in each period in a set number period according to the predicted value and the cumulative probability value of the item sales volume corresponding to each branch point.
3. The method of claim 1, wherein said determining from each of said first cumulative probability distribution data first probability distribution data corresponding to an amount of sales of the item over the period comprises:
and determining a prediction probability value corresponding to the predicted value of the commodity sales volume according to the cumulative probability value of each quantile point based on the conversion relation between the probability distribution and the cumulative probability distribution to obtain first probability distribution data of the commodity sales volume in each period.
4. The method of claim 1, wherein the joint probability distribution based calculation method determining a second cumulative probability distribution data of the predicted total sales of the items for all the set number of cycles from the first probability distribution data for each cycle comprises:
determining second probability distribution data of the total sales of the articles in all the set number periods according to the first probability distribution data based on a calculation method of joint probability distribution;
and determining second cumulative probability distribution data of the corresponding total commodity sales according to the second probability distribution data of the total commodity sales of each set number period based on the conversion relation between the probability distribution and the cumulative probability distribution.
5. The method according to claim 4, wherein the joint probability distribution-based calculation method for determining second probability distribution data of total sales of the items for all the set number of cycles from each of the first probability distribution data includes:
based on a calculation method of joint probability distribution, carrying out weighted calculation on the prediction probability value of the corresponding item sales volume in each first probability distribution data to obtain the probability value of the total sales volume of all the items in a set number period;
and aggregating the probability values with the same total commodity sales in the set number periods to obtain second probability distribution data of the total commodity sales in all the set number periods.
6. The method of claim 2, further comprising:
sampling the predicted value of the commodity sales volume in each period in a set number period for preset times to obtain a sampled commodity sales volume set;
simulating the sampling article sales volume set to obtain a simulation result set;
and determining an empirical quantile corresponding to the simulation result set, and determining a predicted value of the total commodity sales in each set quantity period according to the empirical quantile.
7. The method of claim 1, wherein determining restocking strategy data for the corresponding item according to the predicted value of the total sales volume of the item comprises:
acquiring the current inventory allowance of the article;
and updating the current inventory allowance according to the predicted value of the total item sales in each set quantity period.
8. An apparatus for determining an item restocking strategy based on a predicted total sales volume for a plurality of cycles, the apparatus comprising:
the acquisition module is used for acquiring historical sales volume data of the sales volume of the article;
the prediction module is used for inputting the historical sales data into a quantile prediction model and determining first cumulative probability distribution data of the sales of the articles in each period in a set number period according to quantile parameter values of the quantile prediction model;
a first determining module, configured to determine, according to each of the first cumulative probability distribution data, first probability distribution data corresponding to an amount of sales of the item in the period;
a second determining module, configured to determine, according to the first probability distribution data of each period, second cumulative probability distribution data of the total predicted sales of the items for all the periods of the set number based on a calculation method of a joint probability distribution;
a third determining module, configured to determine a predicted value of the total commodity sales in each of the set number periods according to the second cumulative probability distribution data;
and the fourth determination module is used for determining replenishment strategy data of the corresponding article according to the predicted total sales value of the article.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010742505.0A CN114066491A (en) | 2020-07-29 | 2020-07-29 | Method and device for determining goods replenishment strategy based on multi-cycle total sales forecast |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010742505.0A CN114066491A (en) | 2020-07-29 | 2020-07-29 | Method and device for determining goods replenishment strategy based on multi-cycle total sales forecast |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114066491A true CN114066491A (en) | 2022-02-18 |
Family
ID=80226701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010742505.0A Pending CN114066491A (en) | 2020-07-29 | 2020-07-29 | Method and device for determining goods replenishment strategy based on multi-cycle total sales forecast |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114066491A (en) |
-
2020
- 2020-07-29 CN CN202010742505.0A patent/CN114066491A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200293360A1 (en) | Techniques to manage virtual classes for statistical tests | |
| Aytac et al. | Characterization of demand for short life-cycle technology products | |
| CN112598248B (en) | Load prediction method, load prediction device, computer equipment and storage medium | |
| US20130030785A1 (en) | Computer resource utilization modeling for multiple workloads | |
| Kacar et al. | A comparison of multiple linear regression approaches for fitting clearing functions to empirical data | |
| CN115689018A (en) | Material demand prediction method, device, equipment, storage medium and program product | |
| CN115564152A (en) | Carbon emission prediction method and device based on STIRPAT model | |
| Cobb et al. | Accurate lead time demand modeling and optimal inventory policies in continuous review systems | |
| Yang et al. | Bayesian geoadditive modelling of climate extremes with nonparametric spatially varying temporal effects | |
| CN112686470A (en) | Power grid saturation load prediction method and device and terminal equipment | |
| CN112819507A (en) | Service pushing method and device, electronic equipment and readable storage medium | |
| CN111626789A (en) | House price prediction method, device, equipment and storage medium | |
| Desmet et al. | Safety stock optimisation in two-echelon assembly systems: normal approximation models | |
| CN111768246A (en) | Data processing method, model establishing device and electronic equipment | |
| CN114066491A (en) | Method and device for determining goods replenishment strategy based on multi-cycle total sales forecast | |
| CN115473219A (en) | Load prediction method, load prediction device, computer equipment and storage medium | |
| CN112132498A (en) | Inventory management method, device, equipment and storage medium | |
| CN113837782B (en) | Periodic term parameter optimization method and device of time sequence model and computer equipment | |
| CN114925919A (en) | Service resource processing method and device, computer equipment and storage medium | |
| CN108471362B (en) | Resource allocation prediction technique and device | |
| CN113837467A (en) | Point-to-point energy project evaluation method and device, computer equipment and storage medium | |
| JP2004272674A (en) | Prediction system and prediction method | |
| CN111612419A (en) | Method and device for processing power declaration data and computer equipment | |
| CN110162697A (en) | User browsing behavior analysis method, device, medium and terminal based on big data | |
| CN112749821B (en) | Express delivery quantity prediction method, device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |