CN104820938A - Optimal ordering period prediction method for seasonal and periodic goods - Google Patents

Abstract

The invention discloses an optimal ordering period prediction method for seasonal and periodic goods, and the method is suitable for the goods which is periodic and is sensitive to seasons, wherein the sales volume of the goods increases at first and then decreases. The method comprises the steps: predicting the sales demands of the goods in the current period based on the history sales data of the goods; considering the ordering cost of the goods, the stock cost and the interest cost of a bank loan; and calculating the optimal ordering period of the goods through a quantitative method. Output results enable the decision facilitating the ordering period of the current goods of an enterprise to be made in advance, thereby providing accurate and scientific judgment for the decisions of the requirements for stock capacity, the stock cost control and the amount of bank load of the enterprise.

Description

A kind of optimal replenishment period forecasting method of seasonality, periodically commodity

Technical field

The present invention relates to a kind of optimal replenishment period forecasting method of seasonality, periodically commodity, be applicable to periodic phenomena, to sensitivity in season, within the cycle, sales volume first increases the commodity reduced afterwards, by the prediction in the optimal replenishment cycle to these commodity, be conducive to enterprise and carry out decision-making in advance in the current ordering cycle to commodity, belong to information prediction technical field.

Background technology

The development of market economy exacerbates the competition between enterprise, and competition among enterprises changes Cost Competition into from price competition, inventory cost as its important component part, the effect ever more important in competition among enterprises.Current fixed order quantity model method constraint condition is too much, less consideration actual conditions, are difficult to reflect actual market demand dynamic, complicated and changeable well, cause the frequent generation in unsalable, out of stock phenomenon ground, make turnover rate of materials low, also do not consider operational capital.

Summary of the invention

Goal of the invention: for Problems existing in the prediction of existing goods ordering cycle with not enough, the invention provides a kind of optimal replenishment period forecasting method of seasonality, periodically commodity, based on the historic sales data of commodity, the current sale demand of prediction commodity, and consider that the ordering cost of commodity, stockholding cost and bank loan also cease cost, the optimal replenishment cycle of these commodity is drawn by quantivative approach, thus improve enterprise's commodity turnover quality, save business inventory cost, the enterprise market competitiveness.

Technical scheme: a kind of optimal replenishment period forecasting method of seasonality, periodically commodity, is applicable to periodic phenomena, to sensitivity in season, main manifestations is the commodity that in the cycle, sales volume first increases rear minimizing, such as: cold drink, down jackets etc.Specifically comprise the steps:

(1) obtain the historical sales data on commodity in the sales cycle, sales cycle, usually in units of the moon, is generally 3 to 24 months, according to the sequence of time order and function order, builds time series data { Y _t;

(2) next time period sales cycle offtake data { F is predicted _t, utilize following formula:

S′ _t＝aY _t+(1-a)S′ _t-1

S″ _t＝aS′ _t+(1-a)S″ _t-1

S″′ _t＝aS″ _t+(1-a)S″′ _t-1

a _t＝3S′ _t-3S″ _t+S″′ _t

b _t＝[(6-5a)S′ _t-(10-8a)S″ _t+(4-3a)S″′ _t]*a/(1-a) ²

c _t＝(S′ _t-2S″ _t+S″′ _t)*a/(1-a) ²

F _t+m＝a _t+b _t*m+c _t*m ²/2

Y _tfor the effective sale amount in t period in the upper cycle, F _t+mfor the prediction sales volume apart from t period in the upper period m cycle, S ' _tfor the single exponential smoothing value in t period in predetermined period, S " _tfor the double smoothing value in t period in predetermined period, S " ' _tfor the Three-exponential Smoothing value in t period in predetermined period, S ' _t-1for the single exponential smoothing value in t-1 period in predetermined period, S " _t-1for the double smoothing value in t-1 period in predetermined period, S " ' _t-1for the Three-exponential Smoothing value in t-1 period in predetermined period, a _t, b _t, c _tfor the smoothing factor in t period in predetermined period, a (0<a<1) is smoothing factor, and a value in this method is set by the user, initial value:

S' ₂＝aY ₂+(1-a)Y ₁

S″ ₂＝aS' ₂+(1-a)Y ₁

S″′ ₂＝aS″ ₂+(1-a)Y ₁

F _2+m＝a ₂+b ₂*m+c ₂*m ²/2

Y ₁for the effective sale amount in the 1st period in the upper cycle, Y ₂for the effective sale amount in the 2nd period in the upper cycle, S' ₂for the single exponential smoothing value in the 2nd period in predetermined period, S " ₂for the double smoothing value in the 2nd period in predetermined period, S " ' ₂for the Three-exponential Smoothing value in the 2nd period in predetermined period, a ₂, b ₂, c ₂for the smoothing factor in the 2nd period in predetermined period

(3) the offtake data { F will doped _tfit to curve, obtain the function expression y of this curve _t=a ₁t ⁿ+ a ₂t ^n-1+ a ₃t ^n-2+ ...+a _n-1t ¹+ a _nt ⁰;

T is the t period of predetermined period, y _tfor the prediction sales volume in t period in predetermined period, t ⁰, t ¹t ⁿfor 0 power, 1 power of t numerical value ... n power, a ₁, a ₂a _nfor coefficient.

(4) the sales cycle T of these commodity is obtained according to historical data _n, each ordering cost is K, and the purchase cost of every part commodity is c, and the lease expenses that warehouse is each issue is R, and the volume in warehouse is V, and the volume of every part commodity is v, and the order cycle time is T, then order number of times is T _n/ T, unit commodity stocks cost of carry coefficient is h=v/V*R;

(5) n-th order cycle time T _norder q _nbe the n-th order cycle time T _namount on order, be prediction sales volume y _tfrom the integration of (n-1) T to nT, so the order goods cost of the n-th order cycle time is cQ _n;

(6) n-th cycle T _nthe stockholding cost in warehouse is

(7) each order cycle time, consider that the owner of cargo is because needing to order goods, buy articles from the storeroom, lease warehouse and the M=K+cQ that gets a bank loan _n+ C, loan rate per month is r, repays principal and interest M (1+r) after agreement T time section ^t, as the interest cost of this order cycle time.

(8) according to the data obtained above, the inventory cost can trying to achieve the every first phase order cycle time generation of these commodity is M (1+r) ^t.

(9) then at the sales cycle T of commodity _nin, total inventory cost

$={\mathrm{\&Sigma;}}_{n=1}^{n=\frac{\mathrm{Tn}}{T}}M{(1+r)}^T=[T_{N}k/T+(c+h/2)\underset{0}{\overset{T_N}{\&Integral;}}{y}_t\mathrm{dt}]*{(1+r)}^T;$

(10) according to the equation in step (9), obtain order cycle time T and make within the sales cycle of commodity, total inventory cost is minimum, and calculated order cycle time T is this commodity projection Optimal Order cycle T ^*.

(11) analysis that predicts the outcome exports.

Accompanying drawing explanation

Fig. 1 is the method flow diagram of the embodiment of the present invention;

Fig. 2 is commodity projection sales data scatter diagram and the curve figure of the embodiment of the present invention;

Fig. 3 is total inventory cost predicted data scatter diagram and the curve figure of the embodiment of the present invention.

Embodiment

Below in conjunction with specific embodiment, illustrate the present invention further, these embodiments should be understood only be not used in for illustration of the present invention and limit the scope of the invention, after having read the present invention, the amendment of those skilled in the art to the various equivalent form of value of the present invention has all fallen within the application's claims limited range.

As shown in Figure 1, optimal replenishment cycle T is predicted ^*, comprise the steps:

Exemplary application:

(1) obtain commodity historical sales data, according to the sequence of time order and function order, build time series data { Y _t, as follows:

t	0	1	2	3	4	5	6	7	8	9
											Y t	439	567	765	986	1127	2343	4563	7632	6789	7489
t	10	11	12	13	14	15	16	17	18	19
											Y t	8545	11006	9743	9642	8674	8631	7642	5632	4623	3256
t	20	21	22	23
											Y t	2109	944	524	235

(2) subsequent time period article sales data { F is predicted _t, utilize following formula:

S′ _t＝aY _t+(1-a)S′ _t-1

S″ _t＝aS′ _t+(1-a)S″ _t-1

S″′ _t＝aS″ _t+(1-a)S″′ _t-1

a _t＝3S′ _t-3S″ _t+S″′ _t

b _t＝[(6-5a)S′ _t-(10-8a)S″ _t+(4-3a)S″′ _t]*a/(1-a) ²

c _t＝(S′ _t-2S″ _t+S″′ _t)*a/(1-a) ²

F _t+m＝a _t+b _t*m+c _t*m ²/2

A value in this method is given by user, initial value:

S' ₂＝aY ₂+(1-a)Y ₁

S″ ₂＝aS' ₂+(1-a)Y ₁

S″′ ₂＝aS″ ₂+(1-a)Y ₁

F _2+m＝a ₂+b ₂*m+c ₂*m ²/2

Because predict the data in next cycle, so m=1 herein, setting a=0.4

Draw predicted data { F _t, as follows:

t	0	1	2	3	4	5	6	7	8	9
											F t	439	539	648	999	1375	1574	3459	7195	12298	10691
t	10	11	12	13	14	15	16	17	18	19
											F t	10064	10621	13905	11540	9963	7627	7061	5613	2473	944
t	20	21	22	23
											F t	-533	-1532	-2424	-2106

(3) by dope article sales data { F _tfit to curve, as shown in Figure 2, obtain the function expression of this curve

Y=291.95x-555.96x ²+ 172.34x ³-16.463x ⁴+ 0.63182x ⁵-0.0085895x ⁶+ 839.1, function expression automatically provides after curve.

(4) this merchandise sales cycle T doped is obtained _n=23 months, each ordering cost K=1000 unit, the purchase cost of every part commodity is c=5 unit, the lease expenses of warehouse every day is R=1000 unit, and the volume in warehouse is V=1000 cubic meter, and the volume of every part commodity is v=1 cubic meter, order cycle time is T, then order number of times is T _n/ T, the stockholding cost of unit commodity every days is h=v/V*R=1/1000*1000=1 unit;

(5) n-th order cycle time T _norder so the order goods cost of the n-th order cycle time is ${\mathrm{cQ}}_n=5\underset{(n-1)T}{\overset{\mathrm{nT}}{\&Integral;}}{y}_t\mathrm{dt};$

(6) n-th cycle T _nthe stockholding cost in warehouse is

$C=h/2*\underset{(n-1)T}{\overset{\mathrm{nT}}{\&Integral;}}{y}_t\mathrm{dt}=1/2*\underset{(n-1)T}{\overset{\mathrm{nT}}{\&Integral;}}{y}_t\mathrm{dt};$

(7) each order cycle time, consider that the owner of cargo is because needing order, buying articles from the storeroom, leasing warehouse and get a bank loan $M=K+{\mathrm{cQ}}_n+C=1000+(5+1/2)*\underset{(n-1)T}{\overset{\mathrm{nT}}{\&Integral;}}{y}_t\mathrm{dt}=1000+11/2*\underset{(n-1)T}{\overset{\mathrm{nT}}{\&Integral;}}{y}_t\mathrm{dt},$ Loan rate per month is r=2%, repays principal and interest M (1+r) after agreement T time section ^t, as the interest cost of this order cycle time.

(8) according to the data obtained above, the inventory cost can trying to achieve the every first phase order cycle time generation of these commodity is $M{(1+r)}^T=(1000+11/2*\underset{(n-1)T}{\overset{\mathrm{nT}}{\&Integral;}}{y}_t\mathrm{dt})1.02^T.$

(9) then at the sales cycle T of commodity _nin, total inventory cost

$=[T_{N}k/T+(c+h/2)\underset{0}{\overset{T_N}{\&Integral;}}{y}_t\mathrm{dt}]*{(1+r)}^T=(1000*23/T+11/2*\underset{0}{\overset{23}{\&Integral;}}{y}_t\mathrm{dt}){1.02}^T;$

(10) this commodity projection Optimal Order cycle T is obtained ^*=1.3 months, make within the sales cycle of commodity, total inventory cost is minimum is 678923.9 yuan.Result as shown in Figure 3.

(11) analysis that predicts the outcome exports.This commodity projection Optimal Order cycle T ^*=1.3 months.

Claims (1)

1. an optimal replenishment period forecasting method for seasonality, periodically commodity, is characterized in that, comprise the steps:

(1) obtain the historical sales data on commodity in the sales cycle, according to the sequence of time order and function order, build time series data { Y _t;

(2) subsequent time period offtake data { F is predicted _t, utilize following formula:

S′ _t＝aY _t+(1-a)S′ _t-1

S″ _t＝aS′ _t+(1-a)S″ _t-1

S″′ _t＝aS″ _t+(1-a)S″′ _t-1

a _t＝3S′ _t-3S″ _t+S″′ _t

b _t＝[(6-5a)S′ _t-(10-8a)S″ _t+(4-3a)S″′ _t]*a/(1-a) ²

c _t＝(S′ _t-2S″ _t+S″′ _t)*a/(1-a) ²

F _t+m＝a _t+b _t*m+c _t*m ²/2

Y _tfor the effective sale amount in t period in upper one-period, F _t+mfor the prediction sales volume apart from t period in the upper period m cycle, S ' _tfor the single exponential smoothing value in t period in predetermined period, S " _tfor the double smoothing value in t period in predetermined period, S " ' _tfor the Three-exponential Smoothing value in t period in predetermined period, S ' _t-1for the single exponential smoothing value in t-1 period in predetermined period, S " _t-1for the double smoothing value in t-1 period in predetermined period, S " ' _t-1for the Three-exponential Smoothing value in t-1 period in predetermined period, a _t, b _t, c _tfor the smoothing factor in t period in predetermined period, a (0<a<1) is smoothing factor, and a value in this method is set by the user, initial value:

S′ ₂＝aY ₂+(1-a)Y ₁

S″ ₂＝aS′ ₂+(1-a)Y ₁

S″′ ₂＝aS″ ₂+(1-a)Y ₁

F _2+m＝a ₂+b ₂*m+c ₂*m ²/2

Y ₁for the effective sale amount in the 1st period in the upper cycle, Y ₂for the effective sale amount in the 2nd period in the upper cycle, S ' ₂for the single exponential smoothing value in the 2nd period in predetermined period, S " ₂for the double smoothing value in the 2nd period in predetermined period, S " ' ₂for the Three-exponential Smoothing value in the 2nd period in predetermined period, a ₂, b ₂, c ₂for the smoothing factor in the 2nd period in predetermined period;

(3) the offtake data { F will doped _tfit to curve, obtain the function expression y of this curve _t=a ₁t ⁿ+ a ₂t ^n-1+ a ₃t ^n-2+ ...+a _n-1t ¹+ a _nt ⁰;

T is the t period of predetermined period, y _tfor the prediction sales volume in t period in predetermined period, t ⁰, t ¹t ⁿfor 0 power, 1 power of t numerical value ... n power, a ₁, a ₂a _nfor coefficient.

(4) the sales cycle T of these commodity is obtained according to historical data _n, each ordering cost is K, and the purchase cost of every part commodity is c, and the lease expenses that warehouse is each issue is R, and the volume in warehouse is V, and the volume of every part commodity is v, and the order cycle time is T, then order number of times is T _n/ T, unit commodity stocks cost of carry coefficient is h=v/V*R;

(5) n-th order cycle time T _norder q _nbe the n-th order cycle time T _namount on order, be prediction sales volume y _tfrom the integration of (n-1) T to nT, so the order goods cost of the n-th order cycle time is cQ _n;

(6) n-th cycle T _nthe stockholding cost in warehouse is

(7) each order cycle time, consider that the owner of cargo is because needing to order goods, buy articles from the storeroom, lease warehouse and the M=K+cQ that gets a bank loan _n+ C, loan rate per month is r, repays principal and interest M (1+r) after agreement T time section ^t, as the interest cost of this order cycle time.

(8) according to the data obtained above, the inventory cost can trying to achieve the every first phase order cycle time generation of these commodity is M (1+r) ^t;

(9) then at the sales cycle T of commodity _nin, total inventory cost

$={\mathrm{\&Sigma;}}_{n=1}^{n=\frac{\mathrm{Tn}}{T}}M{(1+r)}^T=[T_{N}k/T+(c+h/2)\underset{0}{\overset{T_N}{\&Integral;}}{y}_t\mathrm{dt}]*{(1+r)}^T;$

(10) according to the equation in step (9), obtain order cycle time T and make within the sales cycle of commodity, total inventory cost is minimum, and calculated order cycle time T is this commodity projection Optimal Order cycle T ^*;

(11) analysis that predicts the outcome exports.