CN116777067A - Dynamic coefficient-based installed prediction algorithm - Google Patents
Dynamic coefficient-based installed prediction algorithm Download PDFInfo
- Publication number
- CN116777067A CN116777067A CN202310758135.3A CN202310758135A CN116777067A CN 116777067 A CN116777067 A CN 116777067A CN 202310758135 A CN202310758135 A CN 202310758135A CN 116777067 A CN116777067 A CN 116777067A
- Authority
- CN
- China
- Prior art keywords
- month
- mds
- day
- total
- dynamic
- 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
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 16
- 238000009434 installation Methods 0.000 claims abstract description 61
- 238000012417 linear regression Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims description 6
- 230000007774 longterm Effects 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000003203 everyday effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Mathematical Analysis (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Human Resources & Organizations (AREA)
- Computational Mathematics (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Algebra (AREA)
- Marketing (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a loading prediction algorithm based on dynamic coefficients, which comprises the following specific steps: s1, acquiring a current date, calculating the difference between the current date and the current month starting date, and calculating by adopting different formulas according to whether the difference between the dates is less than 15 days or not; s2, obtaining each related variable affecting the dynamic coefficient, calculating the proportion of the actual installed number in each time period to the MDS demand number, summing according to different weight ratios and different influence on the month installation according to the proportion, obtaining the dynamic installed coefficient, and S3, constructing a linear regression equation by using the dynamic coefficient, and calculating the expected installed number in the month. The invention effectively solves the problem that the installation demand plan formulated by the MDS table in the prior art can not be adjusted along with the actual situation of the recent and long-term history installation.
Description
Technical Field
The invention relates to the technical field of linear regression prediction, in particular to an installed prediction algorithm based on dynamic coefficients.
Background
Linear regression is a statistical analysis method for determining the quantitative relationship of interdependence between two or more variables by using regression analysis in mathematical statistics, and is widely applied, and the expression form of unitary linear regression is y=w' x+e.
The company makes a daily installation demand plan in 3 months according to the automobile installation demand of a host factory and fills the installation demand plan into an installation demand table (hereinafter referred to as an MDS table), but because various uncertainty factors (such as factors such as adjustment by automobile industry policy, high Wen Xiandian in summer can influence the capacity of a gearbox and an electric drive assembly of the company, and can not guarantee the supply demand of the installation), the installation demand of the MDS table can be different from the actual installation number to a certain extent, the final day of each month can be subjected to centralized installation, the installation number of the day is about twice of the plan, in addition, the automobile industry has a light season and a light season which have obvious difference in the completion rate of the installation plan number of the MDS table, so the installation demand plan of the current MDS table can not be adjusted along with the actual situation of the recent and long-term history installation.
Disclosure of Invention
Aiming at the defects, the invention provides a novel linear prediction algorithm based on dynamic coefficients, and the invention effectively solves the problem that the established installation demand plan can not be adjusted along with the actual conditions of recent and long-term history installation due to the MDS table in the prior art.
The technical scheme for solving the technical problems is as follows:
an installed prediction algorithm based on dynamic coefficients comprises the following steps:
s1, acquiring a current date, calculating the difference between the current date and the current month starting date, and calculating by adopting different formulas according to whether the difference between the dates is less than 15 days or not;
s2, acquiring each related variable affecting the dynamic coefficient, calculating the proportion of the actual installed number in each time period to the MDS demand number, summing according to different weight ratios and different influence on the month installation according to the different influence on the month installation, and obtaining the dynamic installed coefficient, wherein the specific process is carried out by the following formula:
dynamic coefficient ω=0.7×a/b+0.2×m/n+0.1×p/q 15 days before month;
wherein the variable a is the sum of the actual installed numbers of 15 days before the current day, b is the sum of the MDS demand numbers of 15 days before the current day, m is the total installed number of the same month in the last year, n is the total demand number of the MDS of the same month in the last year, p is the total installed number of the same month in the last year, and q is the total demand number of the MDS of the same month in the last year;
the dynamic coefficient ω=0.7×a/c+0.2×m/n+0.1×p/q after 15 days of month;
wherein the variable a is the total installed number from the beginning of the month to the day, c is the total MDS required number from the beginning of the month to the day, m is the total installed number of the same month in the last year, n is the total required number of the MDS in the same month in the last year, p is the total installed number of the same month in the last year, and q is the total required number of the MDS in the same month in the last year;
and S3, constructing a linear regression equation by using dynamic coefficients, and calculating the expected installation number in the current month.
Further, the linear regression equation has dynamic coefficients, and the specific process is performed by the following formula:
15 days before the current month: y=ω x+a;
where ω is the dynamic coefficient of claim 1, x is the sum of the demand for MDS on and after the day 14, a is the sum of the actual installed numbers 15 days before the day;
y=ω x+a+b after 15 days of the month;
where ω is the dynamic coefficient of claim 1, x is the total MDS demand number from the day to the end of the month, a is the total installed number from the beginning of the month to the day before the day, and b is the MDS demand number from the last day of the month.
Further, when the algorithm is calculated at two time periods, 15 days before the month and 15 days after the month, there are different linear regression equations, where the linear regression equation for the 15 days before the month is y=ω x+a, and the linear regression equation after the 15 days of the month is y=ω x+a+b.
The beneficial effects of the invention are as follows: the invention effectively solves the problem that the traditional linear prediction algorithm cannot accurately predict due to the influence of various factors on the installation plan of the enterprise, and achieves the effects of continuously optimizing predicted data by utilizing the recent and long-term historical installation conditions and more accurately predicting data closer to the end of the month.
Drawings
FIG. 1 is a block diagram of a dynamic coefficient-based linear prediction algorithm according to the present invention;
FIG. 2 is a schematic diagram of month installation prediction of the month 11 day;
fig. 3 is a schematic diagram of the month installation prediction situation at 21 st month.
Detailed Description
The invention is further described with reference to the drawings and detailed description.
As shown in fig. 1 to 3, a dynamic coefficient-based installed prediction algorithm includes the steps of:
s1, acquiring a current date, calculating the difference between the current date and the current month starting date, and calculating by adopting different formulas according to whether the difference between the dates is less than 15 days or not;
s2, acquiring each related variable affecting the dynamic coefficient, calculating the proportion of the actual installed number in each time period to the MDS demand number, summing according to different weight ratios and different influence on the month installation according to the different influence on the month installation, and obtaining the dynamic installed coefficient, wherein the specific process is carried out by the following formula:
dynamic coefficient ω=0.7×a/b+0.2×m/n+0.1×p/q 15 days before month;
wherein the variable a is the sum of the actual installed numbers of 15 days before the current day, b is the sum of the MDS demand numbers of 15 days before the current day, m is the total installed number of the same month in the last year, n is the total demand number of the MDS of the same month in the last year, p is the total installed number of the same month in the last year, and q is the total demand number of the MDS of the same month in the last year;
the dynamic coefficient ω=0.7×a/c+0.2×m/n+0.1×p/q after 15 days of month;
wherein the variable a is total installed number before the beginning of month to the day, c is total MDS required number before the beginning of month to the day, m is total installed number of the same month in the last year, n is total MDS required number of the same month in the last year, p is total installed number of the same month in the last year, q is total MDS required number of the same month in the last year;
and S3, constructing a linear regression equation by using dynamic coefficients, and calculating the expected installation number in the current month.
Preferably, the linear regression equation has dynamic coefficients, and the specific process is performed by the following formula:
15 days before the current month: y=ω x+a;
where ω is the dynamic coefficient of claim 1, x is the sum of the demand for MDS on and after the day 14, a is the sum of the actual installed numbers 15 days before the day;
y=ω x+a+b after 15 days of the month;
where ω is the dynamic coefficient of claim 1, x is the total MDS demand number from the day to the end of the month, a is the total installed number from the beginning of the month to the day before the day, and b is the MDS demand number from the last day of the month.
Preferably, when the algorithm is calculated at two time periods, 15 days before the month and 15 days after the month, there are different linear regression equations, the linear regression equation for 15 days before the month is y=ω x+a, and the linear regression equation after 15 days of the month is y=ω x+a+b.
In this embodiment, the linear prediction algorithm based on dynamic coefficients includes the following:
1. related variable definition:
(1) 15 days before the current month:
variable x is the sum of the MDS demand numbers of the day and the following 14 days;
variable a is the sum of the actual installed numbers 15 days before the current day;
variable b, the sum of the 15 day MDS demand numbers before that day;
the variable m is the total assembly machine number of the same month in the last year;
variable n is the MDS total demand number of the same month in the last year;
the variable p is the total assembly machine number of the same month in the previous year;
variable q is the MDS total demand number of the same month in the previous year;
the dynamic installation coefficient omega of the current day can be adjusted together according to different weights according to the installation plan completion rate of the current day 15, the same month installation plan completion rate of the last year and the same month installation plan completion rate of the previous year;
the dynamic loading coefficient of the day is ω=0.7×a/b+0.2×m/n+0.1×p/q.
(2) Stage after day 15 of month:
variable x, total MDS demand number from the current day to the end of the month;
variable a, total installed number before the beginning of month to the day;
variable b, MDS demand number of the last day of the current month;
variable c, total MDS demand number before the beginning of month to the day;
the variable m is the total assembly machine number of the same month in the last year;
variable n is the MDS total demand number of the same month in the last year;
the variable p is the total assembly machine number of the same month in the previous year;
variable q is the MDS total demand number of the same month in the previous year;
the dynamic installation coefficient omega of the current day is adjusted according to different weights according to the installation plan completion rate from the beginning of the month to the current day, the same month installation plan completion rate of the last year and the same month installation plan completion rate of the previous year, and the dynamic installation coefficient omega of the current day is omega=0.7×a/c+0.2×m/n+0.1×p/q.
2. Month estimated installed number calculation logic of each stage:
(1) Stage 15 days before: in the 15 days before month, since the installation situation reference data of the month is less, the installation coefficient omega of the day is calculated according to weights of 0.1, 0.2 and 0.7 respectively by using the previous year installation plan completion rate, the previous year installation plan completion rate and the previous 15 days installation completion rate, thus the installation trend of the day and the following 14 days is judged by the coefficient omega, the calculation formula of omega is omega=0.7 x a/c+0.2 x m/n+0.1 x p/q, the expected installation number of the day and the following 14 days can be obtained by multiplying the MDS demand number of the day and the following 14 days by omega, and the actual installation number of the day and the 15 days before the day can be obtained, and the expected installation number of the month can be obtained by establishing a linear regression equation with dynamic coefficient of y=omega x+a, and the slope omega of the equation can dynamically change every day.
For example: on day 11, the installation prediction situation is shown in fig. 2, where a is the installed number 15 days before the day, b is the MDS demand number 15 days before the day, x is the MDS demand number on the day and 14 days after the day, m is the total installed number of the same month in the last year, n is the total installed number p of the same month in the last year, q is the total installed number of the same month in the last year, the installation coefficient ω=0.7a/b+0.2m/n+0.1p/q of the day, and the predicted installation number y=ω×x+a.
(2) Stage 15 days later: and in the stage after 15 days of the month, calculating the installation coefficient omega of the day by using the same-month installation plan completion rate of the previous year, the same-month installation plan completion rate of the last year and the installation completion rate from the beginning of the month to the current day according to weights of 0.1, 0.2 and 0.7 respectively, judging the installation trend from the current day to the end of the month by using the coefficient omega, wherein the calculation formula of omega is omega=0.7 x a/c+0.2 x p/q, the expected installation number from the current day to the end of the month can be obtained by multiplying omega by the MDS demand number from the current day to the end of the month, and the actual installation number from the beginning of the month to the current day can be obtained by adding the actual installation number before the beginning of the month, and establishing a linear regression equation with dynamic coefficient as y=omega x+a+b, and the slope omega of the equation can dynamically change every day.
For example: on day 21, the installation prediction situation is shown in fig. 3, where a is the total installed number of day 20 before the month, c is the total MDS demand number of day 20 before the month, x is the total MDS demand number from day 21 to the end of the month, m is the total installed number of the same month in the last year, n is the total MDS demand number of the same month in the last year, p is the total installed number of the same month in the last year, q is the total MDS demand number of the same month in the last year, the installation coefficient ω=0.7a/c+0.2m/n+0.1xp/q in the day, and the predicted installed number y=ω×x+a+b.
Finally, it should be explained that: the above embodiments are merely illustrative of the preferred embodiments of the present invention, and not limiting the scope of the present invention; although the invention has been described in detail with reference to the foregoing embodiments, it will be appreciated by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions.
Claims (3)
1. A machine prediction algorithm based on dynamic coefficients, comprising the steps of:
s1, acquiring a current date, calculating the difference between the current date and the current month starting date, and calculating by adopting different formulas according to whether the difference between the dates is less than 15 days or not;
s2, acquiring each related variable affecting the dynamic coefficient, calculating the proportion of the actual installed number in each time period to the MDS demand number, summing according to different weight ratios and different influence on the month installation according to the different influence on the month installation, and obtaining the dynamic installed coefficient, wherein the specific process is carried out by the following formula:
dynamic coefficient ω=0.7×a/b+0.2×m/n+0.1×p/q 15 days before month;
wherein the variable a is the sum of the actual installed numbers of 15 days before the current day, b is the sum of the MDS demand numbers of 15 days before the current day, m is the total installed number of the same month in the last year, n is the total demand number of the MDS of the same month in the last year, p is the total installed number of the same month in the last year, and q is the total demand number of the MDS of the same month in the last year;
the dynamic coefficient ω=0.7×a/c+0.2×m/n+0.1×p/q after 15 days of month;
wherein the variable a is the total installed number from the beginning of the month to the day, c is the total MDS required number from the beginning of the month to the day, m is the total installed number of the same month in the last year, n is the total required number of the MDS in the same month in the last year, p is the total installed number of the same month in the last year, and q is the total required number of the MDS in the same month in the last year;
and S3, constructing a linear regression equation by using dynamic coefficients, and calculating the expected installation number in the current month.
2. The machine prediction algorithm of claim 1, wherein the linear regression equation has dynamic coefficients by the following formula:
15 days before the current month: y=ω x+a;
where ω is the dynamic coefficient of claim 1, x is the sum of the demand for MDS on and after the day 14, a is the sum of the actual installed numbers 15 days before the day;
y=ω x+a+b after 15 days of the month;
where ω is the dynamic coefficient of claim 1, x is the total MDS demand number from the day to the end of the month, a is the total installed number from the beginning of the month to the day before the day, and b is the MDS demand number from the last day of the month.
3. A machine prediction algorithm based on dynamic coefficients according to claim 1, characterized in that when the algorithm is calculated at two time phases, 15 days before the month and 15 days after the month, there are different linear regression equations, the linear regression equation for 15 days before the month is y=ω x+a, and the linear regression equation for 15 days after the month is y=ω x+a+b.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310758135.3A CN116777067A (en) | 2023-06-26 | 2023-06-26 | Dynamic coefficient-based installed prediction algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310758135.3A CN116777067A (en) | 2023-06-26 | 2023-06-26 | Dynamic coefficient-based installed prediction algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116777067A true CN116777067A (en) | 2023-09-19 |
Family
ID=88007722
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310758135.3A Pending CN116777067A (en) | 2023-06-26 | 2023-06-26 | Dynamic coefficient-based installed prediction algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116777067A (en) |
-
2023
- 2023-06-26 CN CN202310758135.3A patent/CN116777067A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Kim et al. | Spare part demand forecasting for consumer goods using installed base information | |
CN111539844B (en) | Moon electricity quantity prediction method based on K-L information quantity and ARIMA error correction | |
CN110941801A (en) | Method and system for predicting reliability of single in-service pressurized water reactor nuclear power unit | |
CN111008727A (en) | Power distribution station load prediction method and device | |
CN116777067A (en) | Dynamic coefficient-based installed prediction algorithm | |
CN113743665A (en) | Power load prediction method based on typical user group of power supply grid | |
JP3641918B2 (en) | Power demand forecast correction method | |
CN117314377A (en) | Human resource management system based on informatization platform | |
Dodson | Determining the optimum schedule for preventive maintenance | |
CN111738507A (en) | Bank clearing position fund payment amount prediction method, device, equipment and medium | |
Tong et al. | Forecasting field failure data for repairable systems using neural networks and SARIMA model | |
US20050096964A1 (en) | Best indicator adaptive forecasting method | |
JPH0922402A (en) | Estimating method for daily maximum power demanded | |
TW202347062A (en) | Method for waiting time prediction in semiconductor factory | |
Murtazaev et al. | The influence of active surface centers on reactivity of mineral additives | |
CN114912700A (en) | Factory workshop electric power energy consumption assessment method and system | |
CN115496522A (en) | Market-oriented trading electric quantity prediction method and system based on moving average | |
CN112100867B (en) | Power grid load prediction method | |
US20130024031A1 (en) | Statistical method to obtain high accuracy in forecasting plant energy use | |
CN101086770A (en) | Accumulation prediction method of spare part requirement | |
CN110633844B (en) | Building energy system simulation prediction method based on EMD and ANN and application | |
CN112330009A (en) | Power grid load prediction method | |
CN112614006A (en) | Load prediction method, device, computer readable storage medium and processor | |
JPH11126102A (en) | Predictive method for demand of gas | |
JP2020102097A (en) | Deterioration level predicting system and deterioration level predicting method |
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 |