CN104732293A - Agricultural product consuming demand predicting method and device - Google Patents

Abstract

The invention provides an agricultural product consuming demand predicting method and device. The method includes the steps that at least one agent is arranged in each map section, and the incidence relation between the agent of the local map section and agents of other map sections is set up; panic parameters of the agents of all the map sections to risk events at the current time point are calculated; according to the corresponding relation between all the panic parameters and consumption quantity reduction parameters, the consumption quantity reduction parameters of the agents of the map sections are determined; consumption reduction parameters of all the agents in the map sections are averaged to obtain a consumption quantity average reduction parameter of a region where the map sections are located. Individual behaviors of consumers and the influence on a market by interaction of all the consumers are simulated through the agents to predict the consuming demand of the region where the map sections are located, and the predicting method well adapts to the change characteristics that the risk events are territorially related, occur suddenly and develop fast.

Description

A kind of Forecasting Methodology of agricultural product consumption demand and device

Technical field

The invention belongs to agricultural product consumption demand prediction field, particularly relate to a kind of Forecasting Methodology and device of agricultural product consumption demand.

Background technology

Food risk weighs the important indicator of Levels of Social Economic Development and consumer's quality of life level.The consumer's risk perception of consumer on food quality and food risk problem, directly affects the buying motive of consumer, and determines purchase intention and the decision behavior of consumer further.In market for farm products, the main aspect of consumer spending demand shift under research risk event is not only in consumer's risk perception, and is the key factor affecting agricultural production order, government decision and whole agricultural product industrial chain profitability.

Under traditional risk perceptions condition, the Forecasting Methodology of agricultural product consumption demand, mainly through econometrics model, carries out the analysis of macroscopic aspect on the various factors affecting the consumption of resident's agricultural product.But these class methods can not adapt to well, and be correlated with, occur suddenly in risk case region, fast-developing Variation Features.

Summary of the invention

Given this, the invention provides a kind of Forecasting Methodology and device of agricultural product consumption demand, be correlated with to adapt to risk case region well, occur suddenly, fast-developing Variation Features.

For achieving the above object, the present invention adopts following technical scheme:

On the one hand, the Forecasting Methodology of a kind of agricultural product consumption demand that the embodiment of the present invention provides, comprising:

Various places figure column is respectively provided to a few proxy server, and sets up the incidence relation between local figure column proxy server and other map column proxy servers; Wherein, described map column is the column that whole distract that risk case relates to divides on map;

According to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, calculates the proxy server of current point in time various places figure column to the panic parameter of described risk case;

Reduce the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely;

In figure column, the consumption figure minimizing parameter of all proxy servers is averaged over the ground, obtains the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location.

Further, described according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, after calculating the panic parameter of the proxy server of current point in time various places figure column, also comprises:

The panic parameter and the consumption figure that obtain consumer reduce parameter;

According to the consumer data obtained, calculate the average panic parameter of consumer;

According to described average panic parameter by the panic standard parameter of each for map column proxy server, obtain the panic parameter of standardization of each proxy server of map column.

Further, reducing the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server also comprises before reducing parameter definitely:

According to described consumer data, the consumption figure calculating panic parameter at different levels reduces level average values, sets up the corresponding relation of panic parameter at different levels and consumption figure minimizing parameter.

Further, described according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, the panic parameter of proxy server to described risk case calculating current point in time various places figure column comprises:

Panic parameter based on the proxy server of following formulae discovery various places figure column:

$P_i(t+1)=\mathrm{\θ}\×{\mathrm{\Σ}}_s\frac{e^{-{\mathrm{\λt}}_s}}{1+e^{\frac{d_{i,s}-b}{a}}}+\frac{1-\mathrm{\θ}}{\left|r\right|}\×{\mathrm{\Σ}}_r\frac{P_r\left(t\right)}{1+e^{\frac{d_{i,r}-b}{a}}}$

Wherein, P _i(t+1) the panic parameter of current agent device i in the t+1 time is represented, behalf risk case, d _i,sthe physical distance between current agent device i to risk case s spot, t _sbe the time that risk case s has occurred, λ is attenuation constant, | r| is the quantity of the contiguous agent device of current agent device i, and r is the sequence number of the contiguous agent of current agent device i, d _i,rfor the physical distance between current agent device i to contiguous agent device r, P _rt () is for proxy server r is in the panic parameter of time t, θ be represent fear that risk case causes proxy server itself input fear with contiguous agent device compared with weight parameter, a and b is the experience factor relevant to Risk of Communication distance, wherein a is that risk underspeeds the position of undergoing mutation by accelerating the slowly distance to risk case spot, and b reduces the distance needed for half for risk.

On the other hand, the prediction unit of a kind of agricultural product consumption demand that the embodiment of the present invention provides, comprising:

Proxy server setting unit, for being respectively provided to a few proxy server on the figure column of various places, and sets up the incidence relation between local figure column proxy server and other map column proxy servers; Wherein, described map column is the column that whole distract that risk case relates to divides on map;

Panic parameter calculation unit, for according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, calculates the proxy server of current point in time various places figure column to the panic parameter of described risk case;

Consumption figure reduces parameter determination unit, and for reducing the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely;

Consumption figure decreased average parameter calculation unit, the consumption figure for all proxy servers in figure column over the ground reduces parameter and averages, and obtains the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location.

Further, described device also comprises:

Consumer data acquiring unit, reduces parameter for the panic parameter and consumption figure obtaining consumer;

Average panic parameter calculation unit, for according to the consumer data obtained, calculates the average panic parameter of consumer;

Panic standard parameter unit, for according to described average panic parameter by the panic standard parameter of each for map column proxy server, obtain the panic parameter of standardization of each proxy server of map column.

Further, described device also comprises:

Corresponding relation sets up unit, for according to described consumer data, calculates the consumption figure minimizing level average values of panic parameter at different levels, sets up the corresponding relation of panic parameter at different levels and consumption figure minimizing parameter.

Compared with prior art, the advantage of technical solution of the present invention is:

The Forecasting Methodology of a kind of agricultural product consumption demand provided by the invention and device, compared with prior art, the present invention is by proxy server phantom shoppers individual behavior and each consumer mutual produced impact on market mutually, obtain the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location, adapt to that be correlated with, occur suddenly in risk case region, fast-developing Variation Features well.

Accompanying drawing explanation

Exemplary embodiment of the present invention will be described in detail by referring to accompanying drawing below, the person of ordinary skill in the art is more clear that above-mentioned and other feature and advantage of the present invention, in accompanying drawing:

The schematic flow sheet of the Forecasting Methodology of the agricultural product consumption demand that Fig. 1 provides for the embodiment of the present invention one;

The association schematic diagram of the various places figure column proxy server that Fig. 2 provides for the embodiment of the present invention one;

The schematic flow sheet of the Forecasting Methodology of the agricultural product consumption demand that Fig. 3 provides for the embodiment of the present invention two;

The structural representation of the prediction unit of the agricultural product consumption demand that Fig. 4 provides for the embodiment of the present invention three;

The structural representation of the prediction unit of the agricultural product consumption demand that Fig. 5 provides for the embodiment of the present invention three.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly, hereinafter with reference to the accompanying drawing in the embodiment of the present invention, by embodiment, technical scheme of the present invention is described clearly and completely, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

Embodiment one

Fig. 1 gives the schematic flow sheet of the Forecasting Methodology of the agricultural product consumption demand that the embodiment of the present invention one provides, the method can be used for the prediction to the consumption demand of agricultural product agricultural product, the method can be performed by the prediction unit of agricultural product consumption demand, and the prediction unit of agricultural product consumption demand can adopt the form of software and/or hardware to realize.As shown in Figure 1, the method comprises:

Step 101, on the figure column of various places, be respectively provided to a few proxy server, and set up the incidence relation between local figure column proxy server and other map column proxy servers.

Wherein, described map column is the column that whole distract that risk case relates to divides on map.Consider the feature that consumer's risk event region is relevant, exemplary, relate to whole multiple columns that area is divided into 50km × 50km on map.Various places figure column is respectively provided to a few proxy server, and each proxy server represents a consumer in real world, is simulated by the behavior pattern of response rule to the entity that it represents preset.The proxy server quantity of various places figure column can the proportional setting of the size of population of column according to the map, and the map column that can be such as 500,000 populations in the ratio of 1:1000 arranges 500 proxy servers.

In addition, the present embodiment also establishes the incidence relation between local figure column proxy server and other map column proxy servers.Each proxy server is connected with multiple proxy server.Exemplary, as shown in Figure 2, each little dotted line frame represents a map column, using A column as local figure column in figure, B column and C column are the map column adjacent with A column, D column is not adjacent with A column, using a proxy server in A column as current agent device, when setting up the incidence relation of a proxy server and other proxy servers, consider the impact of distance on interbehavior, 3 proxy servers be connected with a proxy server are set in A column, 2 proxy servers be connected with a proxy server are altogether set in the column adjacent with A column, with the non-conterminous column of A column, 1 proxy server be connected with a proxy server is set altogether.There is above-mentioned incidence relation in each proxy server for all map columns.By one group autonomous and independently proxy server change respective state by interbehavior, thus realize the impact on other proxy servers.

It should be noted that, the proxy server in the present embodiment can be entity apparatus, also can be software systems.

Step 102, according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, calculates the proxy server of current point in time various places figure column to the panic parameter of risk case.

In step 101, the response rule preset can be the time dependent recursion formula of panic parameter of proxy server; When occurring without any risk case, the initial panic parameter of various places figure column proxy server is 0; When risk case occurs, the change of the panic parameter of each proxy server can follow the response rule of this recursion formula.

In the present embodiment, the panic parameter based on the proxy server of following formulae discovery various places figure column:

$P_i(t+1)=\mathrm{\θ}\×{\mathrm{\Σ}}_s\frac{e^{-{\mathrm{\λt}}_s}}{1+e^{\frac{d_{i,s}-b}{a}}}+\frac{1-\mathrm{\θ}}{\left|r\right|}\×{\mathrm{\Σ}}_r\frac{P_r\left(t\right)}{1+e^{\frac{d_{i,r}-b}{a}}}$

Wherein, P _i(t+1) the panic parameter of current agent device i in the t+1 time is represented, behalf risk case, d _i,sthe physical distance between current agent device i to risk case s spot, t _sbe the time that risk case s has occurred, λ is attenuation constant, and in the present embodiment, λ can get 0.2, | r| is the quantity of the contiguous agent device of current agent device i, and r is the sequence number of the contiguous agent of current agent device i, d _i,rfor the physical distance between current agent device i to contiguous agent device r, P _rt () is for proxy server r is in the panic parameter of time t, θ be represent fear that risk case causes proxy server itself input fear with contiguous agent device compared with weight parameter, in the present embodiment, θ can get 0.8, a and b is the experience factor relevant to Risk of Communication distance, for China, the value of a can be the value of 300km, b can be 800km.

Step 103, reduce the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely.

Wherein, panic parameter can be divided into 0-10 11 grades by panic degree, and the panic degree of the larger expression of numerical value is higher; Consumption figure reduces parameter can represent the number percent that consumption figure reduces.

In step 104, over the ground figure column, the consumption figure of all proxy servers reduces parameter and averages, and obtains the consumption figure decreased average parameter of map column location, with the consumption demand of predictably figure column location.

The Forecasting Methodology of the agricultural product consumption demand that the embodiment of the present invention one provides, by proxy server phantom shoppers individual behavior and each consumer mutual produced impact on market mutually of various places figure column, obtain the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location, adapt to that be correlated with, occur suddenly in risk case region, fast-developing Variation Features well.

Embodiment two

Fig. 3 gives the schematic flow sheet of the Forecasting Methodology of the agricultural product consumption demand that the embodiment of the present invention two provides, the present embodiment is optimized based on above-described embodiment, in the present embodiment, according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, add after calculating the panic parameter of the proxy server of current point in time various places figure column:

Step 303, the panic parameter obtaining consumer and consumption figure reduce parameter.

Exemplary, the present embodiment the mode of questionnaire by inquiry can obtain the panic parameter of consumer and consumption figure reduces parameter, and wherein, questionnaire comprises and reduces parameter for the panic parameter of consumer's risk event consumer and the consumption figure of consumer; Wherein, panic parameter available digital 0-10 represents different panic degree.The present embodiment can be investigated local consumer by providing questionnaire, also can interview by phone consumer by questionnaire according to this.Finally the panic parameter of consumer and consumption figure are reduced parametric statistics in the prediction unit disappearing agricultural product expense demand.

Step 304, according to the consumer data obtained, calculate the average panic parameter of consumer.

After the prediction unit of agricultural product consumption demand obtains above-mentioned consumer data, be simplify and the process that speeds operations, can sample to consumer data, according to the consumer data after sampling, calculate the average panic parameter of consumer.

Step 305, according to the panic standard parameter of average panic parameter by each for map column proxy server, obtain the panic parameter of standardization of each proxy server of map column.

Exemplary, according to average panic parameter, the panic standard parameter of each for map column proxy server is interval to (0,10), obtain the panic parameter of standardization of each proxy server of map column.Wherein, standardization formula is as follows:

$p_i^{*}=\frac{10}{1+e^{-(P_i-\stackrel{\‾}{P})}}$

Wherein P _i ^*the panic parameter of standardization of proxy server i, P _ibe the panic parameter of proxy server i, P is the average panic parameter of all sampling consumers.

Finally, panic for the standardization of proxy server parameter is similar to corresponds to the most close integer.

Further, in the present embodiment, reducing the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server adds before reducing parameter definitely:

Step 306, according to consumer data, the consumption figure calculating panic parameter at different levels reduces level average values, sets up the corresponding relation that panic parameter at different levels and consumption figure reduce parameter.

The Forecasting Methodology of the agricultural product consumption demand that the embodiment of the present invention two provides, by proxy server phantom shoppers individual behavior and each consumer mutual produced impact on market mutually of various places figure column, the corresponding relation of parameter is reduced according to panic parameter at different levels and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely, and then obtain the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location, adapt to risk case region be well correlated with, unexpected generation, fast-developing Variation Features.

Embodiment three

The structural representation of the prediction unit of the agricultural product consumption demand that Fig. 4 provides for the embodiment of the present invention three.As shown in Figure 4, this device comprises:

Proxy server setting unit 40, for being respectively provided to a few proxy server on the figure column of various places, and sets up the incidence relation between local figure column proxy server and other map column proxy servers; Wherein, map column is the column that whole distract that risk case relates to divides on map;

Panic parameter calculation unit 41, for according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, calculates the proxy server of current point in time various places figure column to the panic parameter of risk case;

Consumption figure reduces parameter determination unit 42, and for reducing the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely;

Consumption figure decreased average parameter calculation unit 43, the consumption figure for all proxy servers in figure column over the ground reduces parameter and averages, and obtains the consumption figure decreased average parameter of map column location, with the consumption demand of predictably figure column location.

The prediction unit of the agricultural product consumption demand that the embodiment of the present invention three provides, by analyzing proxy server phantom shoppers individual behavior and each consumer mutual produced impact on market mutually of various places figure column, obtain the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location, adapt to that be correlated with, occur suddenly in risk case region, fast-developing Variation Features well.

Further, with reference to figure 5, the prediction unit of this agricultural product consumption demand also comprises:

Consumer data acquiring unit 50, reduces parameter for the panic parameter and consumption figure obtaining consumer;

Average panic parameter calculation unit 51, for according to the consumer data obtained, calculates the average panic parameter of consumer;

Panic standard parameter unit 52, for according to the panic standard parameter of average panic parameter by each for map column proxy server, obtains the panic parameter of standardization of each proxy server of map column.

Further, the prediction unit of this agricultural product consumption demand also comprises:

Corresponding relation sets up unit 53, for according to described consumer data, calculates the consumption figure minimizing level average values of panic parameter at different levels, sets up the corresponding relation of panic parameter at different levels and consumption figure minimizing parameter.

Note, above are only preferred embodiment of the present invention and institute's application technology principle.Skilled person in the art will appreciate that and the invention is not restricted to specific embodiment described here, various obvious change can be carried out for a person skilled in the art, readjust and substitute and can not protection scope of the present invention be departed from.Therefore, although be described in further detail invention has been by above embodiment, the present invention is not limited only to above embodiment, when not departing from the present invention's design, can also comprise other Equivalent embodiments more, and scope of the present invention is determined by appended right.

Claims (7)

1. a Forecasting Methodology for agricultural product consumption demand, is characterized in that, comprising:

Various places figure column is respectively provided to a few proxy server, and sets up the incidence relation between local figure column proxy server and other map column proxy servers; Wherein, described map column is the column that whole distract that risk case relates to divides on map;

According to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, calculates the proxy server of current point in time various places figure column to the panic parameter of described risk case;

Reduce the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely;

In figure column, the consumption figure minimizing parameter of all proxy servers is averaged over the ground, obtains the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location.

2. method according to claim 1, it is characterized in that, described according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, after calculating the panic parameter of the proxy server of current point in time various places figure column, also comprise:

The panic parameter and the consumption figure that obtain consumer reduce parameter;

According to the consumer data obtained, calculate the average panic parameter of consumer;

According to described average panic parameter by the panic standard parameter of each for map column proxy server, obtain the panic parameter of standardization of each proxy server of map column.

3. method according to claim 2, is characterized in that, reducing the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server also comprises before reducing parameter definitely:

According to described consumer data, the consumption figure calculating panic parameter at different levels reduces level average values, sets up the corresponding relation of panic parameter at different levels and consumption figure minimizing parameter.

4. method according to claim 1, it is characterized in that, described according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, the panic parameter of proxy server to described risk case calculating current point in time various places figure column comprises:

Panic parameter based on the proxy server of following formulae discovery various places figure column:

$P_i(t+1)=\mathrm{\θ}\×{\mathrm{\Σ}}_s\frac{e^{-{\mathrm{\λt}}_s}}{{1+e}^{\frac{d_{i,s}-b}{a}}}+\frac{1-\mathrm{\θ}}{\left|r\right|}\×{\mathrm{\Σ}}_r\frac{P_r\left(t\right)}{{1+e}^{\frac{d_{i,r}-b}{a}}}$

Wherein, P _i(t+1) the panic parameter of current agent device i in the t+1 time is represented, behalf risk case, d _i,sthe physical distance between current agent device i to risk case s spot, t _sbe the time that risk case s has occurred, λ is attenuation constant, | r| is the quantity of the contiguous agent device of current agent device i, and r is the sequence number of the contiguous agent of current agent device i, d _i,rfor the physical distance between current agent device i to contiguous agent device r, P _r(t) for proxy server r is in the panic parameter of time t, θ be represent fear that risk case causes proxy server itself input fear with contiguous agent device compared with weight parameter, a and b is apart from relevant experience factor to Risk of Communication.

5. a prediction unit for agricultural product consumption demand, is characterized in that, comprising:

Proxy server setting unit, for being respectively provided to a few proxy server on the figure column of various places, and sets up the incidence relation between local figure column proxy server and other map column proxy servers; Wherein, described map column is the column that whole distract that risk case relates to divides on map;

Panic parameter calculation unit, for according to the impact of all risk cases on proxy server, and previous time point contiguous agent device is to the panic input parameter of this proxy server, calculates the proxy server of current point in time various places figure column to the panic parameter of described risk case;

Consumption figure reduces parameter determination unit, and for reducing the corresponding relation of parameter according to each panic parameter and consumption figure, the consumption figure of figure column proxy server reduces parameter definitely;

Consumption figure decreased average parameter calculation unit, the consumption figure for all proxy servers in figure column over the ground reduces parameter and averages, and obtains the consumption figure decreased average parameter of map column location, to predict the consumption demand of described map column location.

6. device according to claim 5, is characterized in that, described device also comprises:

Consumer data acquiring unit, reduces parameter for the panic parameter and consumption figure obtaining consumer;

Average panic parameter calculation unit, for according to the consumer data obtained, calculates the average panic parameter of consumer;

Panic standard parameter unit, for according to described average panic parameter by the panic standard parameter of each for map column proxy server, obtain the panic parameter of standardization of each proxy server of map column.

7. device according to claim 6, is characterized in that, described device also comprises:

Corresponding relation sets up unit, for according to described consumer data, calculates the consumption figure minimizing level average values of panic parameter at different levels, sets up the corresponding relation of panic parameter at different levels and consumption figure minimizing parameter.