CN117575118B - Method for planning visit route of science and technology museum - Google Patents
Method for planning visit route of science and technology museum Download PDFInfo
- Publication number
- CN117575118B CN117575118B CN202311429679.1A CN202311429679A CN117575118B CN 117575118 B CN117575118 B CN 117575118B CN 202311429679 A CN202311429679 A CN 202311429679A CN 117575118 B CN117575118 B CN 117575118B
- Authority
- CN
- China
- Prior art keywords
- data
- science
- people flow
- museum
- follows
- 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.)
- Active
Links
- 238000005516 engineering process Methods 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000005457 optimization Methods 0.000 claims abstract description 16
- 230000000694 effects Effects 0.000 claims abstract description 7
- 230000000737 periodic effect Effects 0.000 claims abstract 2
- 230000006870 function Effects 0.000 claims description 37
- 230000008569 process Effects 0.000 claims description 31
- 241000255581 Drosophila <fruit fly, genus> Species 0.000 claims description 19
- 238000013528 artificial neural network Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012417 linear regression Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 10
- 230000007717 exclusion Effects 0.000 claims description 6
- 238000005429 filling process Methods 0.000 claims description 6
- 230000006872 improvement Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000002354 daily effect Effects 0.000 claims description 4
- 241000255588 Tephritidae Species 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000013499 data model Methods 0.000 claims description 3
- 230000003203 everyday effect Effects 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 2
- 230000001360 synchronised effect Effects 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims 1
- 230000001932 seasonal effect Effects 0.000 abstract 1
- 230000008859 change Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram 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"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2131—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on a transform domain processing, e.g. wavelet transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for planning a visit route of a science and technology museum, which comprises a data unit, a science and technology museum people flow prediction unit, an intelligent planning data unit and an intelligent planning model; the prediction data participation unit comprises seasonal factors, periodic factors and historical people flow data, and is used as a people flow prediction basis; the science and technology museum people flow prediction unit comprises a people flow prediction model and an improved intelligent optimization algorithm, and model parameters are optimized through the improved intelligent optimization algorithm, so that the accuracy of model prediction is improved; the intelligent planning data unit comprises people flow prediction data, areas of each sub-museum of the science and technology museum and optimal visit time of each sub-museum, the intelligent planning data are input into an intelligent planning model, a planning target is set, a plurality of scheme routes which do not conflict with each other are obtained, the scheme routes are sent to appointed staff, and the staff guides visitors to visit according to the scheme routes, so that orderly and efficient visit activities are ensured.
Description
Technical Field
The invention relates to the technical field of people flow planning, in particular to a method for planning a visit route of a science and technology center.
Background
With the continuous progress of science and technology, a science and technology museum is used as a place for displaying scientific and technological achievements and popularizing scientific knowledge, however, due to the factors of the great content of exhibition, the flowing of many tourists and the like in the existing science and technology museum, it is often difficult for visitors to efficiently visit various exhibitions in a limited time.
Therefore, a route planning method is always needed, the daily people flow is predicted, the current passenger flow condition is known, and planning and arrangement are performed in advance; the method comprises the steps of obtaining a plurality of mutually non-conflicting route schemes, reducing queuing waiting time of tourists, improving visitor visiting experience, fully utilizing science and technology museum resources, guaranteeing that the overall utilization rate of the science and technology museum in unit time is highest, enabling the science and technology museum to stably and efficiently operate, and improving income of the science and technology museum
In special periods such as holidays, the traffic of science and technology museums is large, so that the problems of crowding of science and technology museums, long waiting time of tourists for visiting, low visiting efficiency and the like are often caused, and the problems lead to holidays with more profit and unsatisfactory final income.
Disclosure of Invention
The invention aims to: the invention aims to reasonably predict the current people flow of a science and technology museum and design a plurality of different non-conflicting visit route schemes according to the predicted current people flow, thereby ensuring the highest utilization rate in the unit time of the science and technology museum and simultaneously minimizing the waiting time in the visit activities of tourists.
The technical scheme is as follows: the invention provides a method for planning a visit route of a science and technology museum, which comprises a data unit, a science and technology museum people flow prediction unit, an intelligent planning data unit, an intelligent planning model and a scheme distribution unit;
Further, the data unit includes four seasons, weather, periodicity, and historical data traffic. The four-season factors and the weather factors comprise spring, summer, autumn and winter, and the corresponding values are respectively set to be 1, 0.75, 1 and 0.6; the weather conditions are also included, wherein the weather conditions comprise sunny days, cloudy days and rainy days, and the corresponding values are 1, 0.6 and 0.2 respectively; the periodicity factors comprise holiday conditions, wherein the values corresponding to Saturday, sunday and holiday are 1, and the value corresponding to working day is 0.3; the historical people flow data are historical people flow data of a science and technology center every day and historical people flow data of each time period within the last two years. The data unit inputs the collected data into a people flow prediction model to predict the daily people flow, and simultaneously inputs the current historical people flow data into a historical people flow database as historical data again.
Further, the science and technology museum people flow prediction unit comprises a people flow prediction model and an intelligent optimization algorithm, and a final prediction result is input into an intelligent planning data unit; the human flow prediction model is an MLP-IWNN combined prediction model combining a multiple linear regression prediction model and a wavelet neural network prediction model, and the implementation process is as follows:
31 A MLP multiple linear regression prediction model is established, and the general form of the model is as follows:
y0＝b0+b1x1i+b2x2i+…+bmxmi+εi,i＝1,2,……,n,
wherein y 0 represents a predicted value of the mass flow of people in a science and technology museum of a multiple linear regression prediction model; b 0、bjb0, representing model regression coefficients, j=1, 2, … …, m; epsilon i represents a random variable generated by y i, in addition to the effect of the independent variable x j (j=1, 2,., m), called random error; x 1i,x2i…xmi respectively represents the spring, summer, autumn and winter, weather factors, periodicity factors and historical people flow data, and the value of m is 4;
32 Calculating a random error, the calculation formula is as follows:
33 A wavelet neural network prediction model is constructed, a wavelet basis function is selected, and the wavelet basis function formula is shown as follows:
34 An input layer, an hidden layer and an output layer of the wavelet neural network are constructed, and the output formula of the hidden layer is as follows:
Wherein w ij is a weight coefficient between an input layer and an hidden layer, a j is a scaling factor of a wavelet basis function, b j is a translation factor, and h j is a hidden layer neuron output;
35 The output layer construction formula is as follows:
Wherein w jk is a weight coefficient between an implicit layer and an output layer, the coefficient is continuously updated in the network training process, and y k is an output value of a kth neuron of the wavelet neural network;
36 A) calculating a network error index function, the calculating function being as follows:
Wherein y' k represents a true value, y k represents a wavelet neural network predicted value, and E represents an error index function for measuring the deviation degree between the true value and the error value;
37 Calculating the variation of the parameters to be adjusted after the d+1st training AndThe calculation formula is as follows:
Wherein eta represents the set learning rate, E is an error index function;
38 Updating weights and wavelet factors by adopting an error back propagation algorithm, wherein the updating formula is as follows:
Wherein d represents the algebra when the parameter is updated, />Indicating the amount of change in the parameters to be adjusted after the d+1st training.
39 The updated formula in step 36) is modified as follows:
Where α is a constant between [0,1 ].
310 Calculating a final science and technology museum people flow predicted value Yl, wherein a calculation formula is as follows:
Yl＝λ1y0+λ2yk
λ1+λ2＝1
Wherein λ 1 and λ 2 represent weight factors, y 0 represents the prediction result of the MLP prediction model, and y k represents the prediction result of the IWNN prediction model;
Further, the method is characterized in that the improved drosophila optimization algorithm EBFOA is adopted to optimize the values of lambda 1 and lambda 2, so that the prediction accuracy is improved, and the specific implementation process is as follows:
The optimization objective function is as follows:
Wherein E represents a precision index, n represents the total number of data amounts, Y represents a real historical human flow data value, and Yl represents a predicted historical human flow data value;
41 Initializing a drosophila population size Sizepop, a maximum number of iterations Maxgen, and drosophila population positions X axis and Y axis;
42 A random search direction and distance are given to the drosophila individuals, and the calculation formula is as follows:
Wherein, the random variable is represented, and the value range is between [ -1,1 ];
43 Optimization and improvement of search step size):
R＝α×e-(β×g)/Margen；
wherein alpha is a step control factor, beta is an index regulation factor, g is the current iteration number, and Maxge is the maximum iteration number;
44 Fruit fly individual update location after improvement):
45 Calculating the distance Dist from the drosophila individual to the origin:
46 A taste concentration determination value S i is calculated, and the calculation formula is as follows:
Si＝1/Disti；
47 Optimizing and improving the taste concentration determination value S i by adopting a sign function:
Si＝sign(2×rand-1)/Dist；
wherein rand is a random number ranging between [0,1] uniformly distributed.
48 Inputting the concentration determination value into an objective function, and calculating a taste concentration value Smell i, wherein the calculation formula is as follows:
Smelli＝Fitness(Si)；
wherein Fitness represents the objective function of calculating the taste intensity value.
49 Obtaining the Drosophila individual with the optimal taste concentration value, and recording the position information and the response taste concentration value, wherein the formula is as follows:
[bestSmell,bestindex]＝min(Smell)；
410 Preserving the optimal taste intensity values bestSmell, performing a location update to form a new population center:
Smellbest＝bestSmell；
411 And (3) iteratively optimizing until the maximum iteration times are met, outputting optimal lambda 1 and lambda 2 values, and inputting the optimal lambda 1 and lambda 2 values into a model for operation to obtain a predicted value with highest precision.
Further, the intelligent planning data unit comprises people flow prediction data, area data of each sub-museum of the science and technology museum and optimal visit time of each sub-museum; the people flow prediction data are obtained through the people flow prediction model; the area of each sub-hall and the optimal visit time of each sub-hall are set fixed values; and inputting the data in the intelligent planning data model into the intelligent planning model to obtain mutually non-conflicting visit routes of the science and technology museum, and sending different visit routes to different staff.
Furthermore, the intelligent scale model planning targets are the conflict-free visit condition of each sub-gallery and the lowest visit waiting time of tourists, the area of each sub-gallery corresponds to the capacity of the total buffer area of the model, the buffer area filling process corresponds to the visitor continuously entering the science and technology gallery, and the buffer area idling process corresponds to the staff taking the tourists to visit. Through the model, a scheme of a non-used visit route can be reasonably planned, workers can lead tourists to visit a science and technology museum according to the scheme, and the visit efficiency of the science and technology museum is improved.
The beneficial effects are that:
1. According to the invention, the characteristics of the traffic of the science and technology museum are combined, the multivariate linear regression prediction model is combined with the wavelet neural network prediction model, the accuracy of the traffic prediction of the science and technology museum is effectively improved, and the wavelet neural network prediction model is optimized by adopting an error back propagation algorithm, so that the error between the traffic of the science and technology museum and the actual traffic is minimized, and therefore, a route is reasonably planned, and congestion is avoided.
2. According to the invention, the improved drosophila optimization algorithm EBFOA with the highest adaptation degree with the science and technology museum people flow prediction model is selected to distribute the weights of the multiple linear regression prediction model and the wavelet neural network prediction model, so that the optimal weight is obtained, and the accuracy of science and technology museum people flow prediction is further improved.
3. The intelligent planning model adopted by the invention acquires different technological museum visit route schemes, ensures that the utilization rate of the technological museum in unit time reaches the highest while the waiting time of tourist visit activities is minimum, improves the operation efficiency of the technological museum and improves the income of the technological museum.
Drawings
FIG. 1 is a structural frame diagram of the present invention
FIG. 2 is a flowchart of the algorithm optimization of the present invention
Detailed Description
Step one: the method comprises the steps that a science and technology museum people flow prediction model obtains prediction data from a prediction data unit, the prediction data comprise four seasons factors, weather factors, periodicity factors and historical people data flow, the four seasons factors and the weather factors comprise spring, summer, autumn and winter, and corresponding values are respectively set to be 1, 0.75, 1 and 0.6; the weather conditions are also included, wherein the weather conditions comprise sunny days, cloudy days and rainy days, and the corresponding values are 1, 0.6 and 0.2 respectively; the periodicity factors comprise holiday conditions, wherein the values corresponding to Saturday, sunday and holiday are 1, and the value corresponding to working day is 0.3; the historical people flow data are historical people flow data of a science and technology center every day and historical people flow data of each time period within the last two years, the data unit inputs the collected data into a people flow prediction model to predict the daily people flow, and meanwhile the current historical people flow data are used as historical data again and are input into a historical people flow database.
Step two: constructing a science and technology museum people flow prediction unit; the people flow prediction model is an MLP-IWNN combined prediction model combining a multiple linear regression prediction model and a wavelet neural network prediction model;
Step three: the drosophila optimization algorithm EBFOA is improved, the values of lambda 1 and lambda 2 are optimized, the prediction accuracy is improved, and the specific implementation process is as follows:
The optimization objective function is as follows:
Wherein E represents a precision index, n represents the total number of data amounts, Y represents a real historical human flow data value, and Yl represents a predicted historical human flow data value;
41 Initializing drosophila population size Sizepop, maximum number of iterations Maxgen, and drosophila population positions X axis and Y axis.
42 A random search direction and distance are given to the drosophila individuals, and the calculation formula is as follows:
Wherein, the random variable is represented, and the value range is between [ -1,1 ];
43 Optimization and improvement of search step size):
R＝α×e-(β×g)/Margen
wherein alpha is a step control factor, beta is an exponential regulation factor, g is the current iteration number, and Maxge is the maximum iteration number.
44 Fruit fly individual update location after improvement):
45 Calculating the distance Dist from the drosophila individual to the origin:
46 A taste concentration determination value S i is calculated, and the calculation formula is as follows:
Si＝1/Disti
47 Optimizing and improving the taste concentration determination value S i by adopting a sign function:
Si＝sign(2×rand-1)/Dist
wherein rand is a random number ranging between [0,1] uniformly distributed.
48 Inputting the concentration determination value into an objective function, and calculating a taste concentration value Smell i, wherein the calculation formula is as follows:
Smelli＝Fitness(Si)
wherein Fitness represents the objective function of calculating the taste intensity value.
49 Obtaining the Drosophila individual with the optimal taste concentration value, and recording the position information and the response taste concentration value, wherein the formula is as follows:
[bestSmell,bestindex]＝min(Smell)
410 Preserving the optimal taste intensity values bestSmell, performing a location update to form a new population center:
Smellbest＝bestSmell
411 And (3) iteratively optimizing until the maximum iteration times are met, outputting optimal lambda 1 and lambda 2 values, and inputting the optimal lambda 1 and lambda 2 values into a model for operation to obtain a predicted value with highest precision.
Step four: the MLP-IWNN combined prediction model combining the multiple linear regression prediction model and the wavelet neural network prediction model is constructed, and the implementation process is as follows:
31 A MLP multiple linear regression prediction model is established, and the general form of the model is as follows:
y0＝b0+b1x1i+b2x2i+...+bmxmi+εi,i＝1,2,……,n,
wherein y 0 represents a predicted value of the mass flow of people in a science and technology museum of a multiple linear regression prediction model; b 0、bjb0, representing model regression coefficients, j=1, 2, … …, m; epsilon i represents a random variable generated by y i, in addition to the effect of the independent variable x j (j=1, 2,., m), called random error; x 1i,x2i…xmi respectively represents the spring, summer, autumn and winter, weather factors, periodicity factors and historical people flow data, and the value of m is 4;
32 Calculating a random error, the calculation formula is as follows:
33 A wavelet neural network prediction model is constructed, a wavelet basis function is selected, and the wavelet basis function formula is shown as follows:
34 An input layer, an hidden layer and an output layer of the wavelet neural network are constructed, and the output formula of the hidden layer is as follows:
wherein w ij is a weight coefficient between an input layer and an hidden layer, a j is a scaling factor of a wavelet basis function, b j is a translation factor, and h j is a hidden layer neuron output; x 1i,x2i…xmi is the above-mentioned spring, summer, autumn and winter, weather factors, periodicity factors and historical people flow data;
35 The output layer construction formula is as follows:
Wherein w jk is a weight coefficient between an implicit layer and an output layer, the coefficient is continuously updated in the network training process, and y k is an output value of a kth neuron of the wavelet neural network;
36 A) calculating a network error index function, the calculating function being as follows:
Wherein y' k represents a true value, y k represents a wavelet neural network predicted value, and E represents an error index function for measuring the deviation degree between the true value and the error value;
37 Calculating the variation of the parameters to be adjusted after the d+1st training AndThe calculation formula is as follows:
Wherein eta represents the set learning rate, E is an error index function;
38 Updating weights and wavelet factors by adopting an error back propagation algorithm, wherein the updating formula is as follows:
Wherein d represents the algebra when the parameter is updated, />Indicating the amount of change in the parameters to be adjusted after the d+1st training.
39 The updated formula in step 36) is modified as follows:
Where α is a constant between [0,1 ].
310 Calculating a final science and technology museum people flow predicted value Yl, wherein a calculation formula is as follows:
Yl＝λ1y0+λ2yk
λ1+λ2＝1
Wherein λ 1 and λ 2 represent weight factors, y 0 represents the prediction result of the MLP prediction model, and y k represents the prediction result of the IWNN prediction model;
Step five: constructing an intelligent planning data unit, wherein the intelligent planning data unit comprises people flow prediction data, area data of each sub-museum of a science and technology museum and optimal visit time of each sub-museum; the people flow prediction data are obtained through a people flow prediction model; the area of each sub-hall and the optimal visit time of each sub-hall are set fixed values; and inputting the data in the intelligent planning data model into the intelligent planning model to obtain mutually non-conflicting visit routes of the science and technology museum, and sending different visit routes to different staff.
Step six: an intelligent planning model is built, the planning targets are the conflict-free visit condition of each sub-gallery and the minimum visit waiting time of tourists, the area of each sub-gallery corresponds to the capacity of a total buffer zone of the model, the buffer zone filling process corresponds to the visitor continuously entering the science and technology gallery, and the buffer zone idling process corresponds to the staff for guiding the tourists to visit. Through the model, a scheme of a non-used visit route can be reasonably planned, workers can lead tourists to visit a science and technology museum according to the scheme, and the visit efficiency of the science and technology museum is improved.
The intelligent planning model is established as follows:
1. Defining the capacity of a buffer zone, wherein the capacity is set as the sum of time spent for visiting all science and technology museums;
2. creating a filling buffer area process and an idle buffer area process, wherein the filling buffer area process refers to a process that tourists enter a science and technology museum and occupy the science and technology museum; the process of idling the buffer zone refers to the process of ending visitor visit and enabling the science and technology museum to be in an idle state;
3. continuously generating data in the process of filling the buffer area, and filling the generated data into the buffer area, wherein if the buffer area is full, waiting is needed until the buffer area has an idle position;
4. The idle buffer process is represented by time, each data has a corresponding service life, when the service life value is zero, the data gives up the occupied buffer, and the process corresponds to the process that when a visiting science and technology museum is in a visiting best time, staff needs to guide tourists away from the museum;
5. the direct mutual exclusion function of data is realized, namely different data cannot occupy the same buffer area at the same time, a mutual exclusion variable Mutex is defined in the process, and when the existing data occupy a certain buffer area, the mutual exclusion variable of the buffer area is 1; when the buffer is idle, the mutex variable of the buffer is 0; when the mutex variable of the buffer is 0, the data can be filled into the buffer;
6. Realizing synchronous communication function, setting a semaphore semaphore, when the service life of data is reduced to 0, sending a signal to a buffer filling process, and filling data into an idle buffer when the buffer filling process receives the signal;
7. and outputting a path planning scheme, namely outputting the reference sequence of each sub-gallery, namely the final visit path after the buffer area is filled.
Claims (6)
1. A method for planning a visit route of a science and technology museum, comprising:
the data unit is used for collecting historical people flow data, wherein 70% of the historical people flow data is used as a training set, 30% of the historical people flow data is used as a testing set, and data support is provided for the science and technology museum people flow prediction unit;
the science and technology museum people flow prediction unit is composed of a people flow prediction model and an intelligent optimization algorithm; the people flow prediction model predicts the people flow according to the data provided by the data unit to obtain historical people flow data of a future day; the intelligent optimization algorithm optimizes parameters of a prediction model, so that the accuracy of prediction is improved, and the people flow prediction model is an MLP-IWNN combined prediction model formed by combining a multiple linear regression prediction model and a wavelet neural network prediction model;
and calculating a final science and technology museum people flow forecast value Yl, wherein a calculation formula is as follows:
Yl＝λ1y0+λ2yk
λ1+λ2＝1
Wherein λ 1 and λ 2 represent weight factors, y 0 represents the prediction result of the MLP prediction model, and y k represents the prediction result of the IWNN prediction model;
The improved drosophila optimization algorithm EBFOA is adopted to optimize the values of lambda 1 and lambda 2, so that the accuracy of prediction is improved;
the intelligent planning data unit is used for storing data, receiving the prediction result of the stadium people flow prediction unit, integrating the prediction result with the area of each stadium of the stadium and the data of the optimal visit time of each stadium, and transmitting the integrated data into the intelligent planning model;
The intelligent planning model carries out route planning by taking no conflict and minimum visit waiting time of each sub-museum as targets according to the data provided by the intelligent planning data unit, outputs a plurality of mutually non-conflicting visit routes and sends the visit routes to the scheme distribution unit;
the scheme distribution unit is composed of a plurality of staff, the staff takes a specified number of tourists to carry out visiting activities according to a route formulated by the intelligent planning model, and the visiting efficiency of the science and technology museums is improved.
2. A method of planning a visit route in a science and technology museum as claimed in claim 1, wherein the data unit includes four seasons, weather and periodicity factors as influencing factors, the four seasons, weather and periodicity factors including spring, summer, autumn and winter; the weather conditions comprise sunny days, overcast days and rainy days, and the periodic factors comprise holiday conditions; the historical people flow data are historical people flow data of a science and technology museum every day and historical people flow data of each time period in the last two years, the data unit inputs collected data into a people flow prediction model of a science and technology museum people flow prediction unit to predict the daily people flow, and meanwhile the historical people flow data of the same day are used as the historical data again and are input into the data unit.
3. A method of planning a visit route in a science and technology museum according to claim 2, wherein the people flow prediction model of the science and technology museum people flow prediction unit inputs the final prediction result into the intelligent planning data unit; the MLP-IWNN combined prediction model is realized as follows:
31 A MLP multiple linear regression prediction model is established, and the general form of the model is as follows:
y0＝b0+b1x1i+b2x2i+...+bmxmi+εi,i＝1,2,……,n,
Wherein y 0 represents a predicted value of the mass flow of people in a science and technology museum of a multiple linear regression prediction model; b 0、bj, representing model regression coefficients, j=1, 2, … …, m; epsilon i represents the random variable generated by y i, in addition to the effect of the argument x j, called random error; x 1i,x2i…xmi respectively represents the spring, summer, autumn and winter, weather factors, periodicity factors and historical people flow data, and the value of m is 4;
32 Calculating a random error, the calculation formula is as follows:
33 A wavelet neural network prediction model is constructed, a wavelet basis function is selected, and the wavelet basis function formula is shown as follows:
34 An input layer, an hidden layer and an output layer of the wavelet neural network are constructed, and the output formula of the hidden layer is as follows:
Wherein w ij is a weight coefficient between an input layer and an hidden layer, a j is a scaling factor of a wavelet basis function, b j is a translation factor, and h j is a hidden layer neuron output;
35 The output layer construction formula is as follows:
Wherein w jk is a weight coefficient between an implicit layer and an output layer, the coefficient is continuously updated in the network training process, and y k is an output value of a kth neuron of the wavelet neural network;
36 A) calculating a network error index function, the calculating function being as follows:
Wherein y' k represents a true value, y k represents a wavelet neural network predicted value, and E represents an error index function for measuring the deviation degree between the true value and the error value;
37 Calculating the variation of the parameters to be adjusted after the d+1st training />The calculation formula is as follows:
Wherein eta represents the set learning rate, E is an error index function;
38 Updating weights and wavelet factors by adopting an error back propagation algorithm, wherein the updating formula is as follows:
Wherein d represents the algebra when the parameter is updated, />Indicating the variation of the parameters to be adjusted after the d+1st training;
39 The updated formula in step 36) is modified as follows:
Where α is a constant between [0,1 ].
4. The method of claim 1, wherein optimizing the values of λ 1 and λ 2 is performed as follows:
The optimization objective function is as follows:
Wherein E represents a precision index, n represents the total number of data amounts, Y represents a real historical human flow data value, and Yl represents a predicted historical human flow data value;
41 Initializing a drosophila population size Sizepop, a maximum number of iterations Maxgen, and drosophila population positions X axis and Y axis;
42 A random search direction and distance are given to the drosophila individuals, and the calculation formula is as follows:
Wherein, the random variable is represented, and the value range is between [ -1,1 ];
43 Optimization and improvement of search step size):
R＝α×e-(β×g)/Margen；
wherein alpha is a step control factor, beta is an index regulation factor, g is the current iteration number, and Maxge is the maximum iteration number;
44 Fruit fly individual update location after improvement):
45 Calculating the distance Dist from the drosophila individual to the origin:
46 A taste concentration determination value S i is calculated, and the calculation formula is as follows:
Si＝1/Disti；
47 Optimizing and improving the taste concentration determination value S i by adopting a sign function:
Si＝sign(2×rand-1)/Dist；
wherein, rand is a random number uniformly distributed in the range of 0, 1;
48 Inputting the concentration determination value into an objective function, and calculating a taste concentration value Smell i, wherein the calculation formula is as follows:
Smelli＝Fitness(Si)；
wherein Fitness represents an objective function for calculating a taste intensity value;
49 Obtaining the Drosophila individual with the optimal taste concentration value, and recording the position information and the response taste concentration value, wherein the formula is as follows:
[bestSmell,bestindex]＝min(Smell)；
410 Preserving the optimal taste intensity values bestSmell, performing a location update to form a new population center:
Smellbest＝bestSmell；
411 And (3) iteratively optimizing until the maximum iteration times are met, outputting optimal lambda 1 and lambda 2 values, and inputting the optimal lambda 1 and lambda 2 values into a model for operation to obtain a predicted value with highest precision.
5. A method of planning a visit route in a science and technology museum as claimed in claim 1, wherein the intelligent planning data unit includes traffic prediction data, area data of each sub-museum of the science and technology museum and optimal visit time of each sub-museum; the people flow prediction data are obtained through a people flow prediction model; the area of each sub-hall and the optimal visit time of each sub-hall are set fixed values; and inputting the data in the intelligent planning data model into the intelligent planning model to obtain mutually non-conflicting visit routes of the science and technology museum, and sending different visit routes to different staff.
6. A method of planning a visit route in a science and technology museum as claimed in claim 1, wherein the intelligent planning model is constructed as follows:
1) Defining the buffer capacity, the capacity is set as the sum of the time spent visiting all science and technology museums,
2) Creating a filling buffer area process and an idle buffer area process, wherein the filling buffer area process refers to a process that tourists enter a science and technology museum and occupy the science and technology museum; the process of idling the buffer zone refers to the process of ending visitor visit and enabling the science and technology museum to be in an idle state;
3) Continuously generating data in the process of filling the buffer area, and filling the generated data into the buffer area, wherein if the buffer area is full, waiting is needed until the buffer area has an idle position;
4) The idle buffer process is represented by time, each data has a corresponding service life, when the service life value is zero, the data gives up the occupied buffer, and the process corresponds to the process that when a visiting science and technology museum is in a visiting best time, staff needs to guide tourists away from the museum;
5) The direct mutual exclusion function of data is realized, namely different data cannot occupy the same buffer area at the same time, a mutual exclusion variable Mutex is defined in the process, and when the existing data occupy a certain buffer area, the mutual exclusion variable of the buffer area is 1; when the buffer is idle, the mutex variable of the buffer is 0; when the mutex variable of the buffer is 0, the data can be filled into the buffer;
6) Realizing synchronous communication function, setting a semaphore semaphore, when the service life of data is reduced to 0, sending a signal to a buffer filling process, and filling data into an idle buffer when the buffer filling process receives the signal;
7) And outputting a path planning scheme, namely outputting the reference sequence of each sub-gallery, namely the final visit path after the buffer area is filled.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311429679.1A CN117575118B (en) | 2023-10-30 | 2023-10-30 | Method for planning visit route of science and technology museum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311429679.1A CN117575118B (en) | 2023-10-30 | 2023-10-30 | Method for planning visit route of science and technology museum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117575118A CN117575118A (en) | 2024-02-20 |
| CN117575118B true CN117575118B (en) | 2024-05-28 |
Family
ID=89894470
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311429679.1A Active CN117575118B (en) | 2023-10-30 | 2023-10-30 | Method for planning visit route of science and technology museum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117575118B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103376116A (en) * | 2012-04-16 | 2013-10-30 | 北京四维图新科技股份有限公司 | Scenic route planning in vehicle navigation |
| CN105046045A (en) * | 2015-06-05 | 2015-11-11 | 河海大学 | Bayesian combination-based Web service QoS prediction method |
| CN105222782A (en) * | 2015-09-07 | 2016-01-06 | 广东欧珀移动通信有限公司 | A kind of route recommendation method and mobile terminal |
| CN106372731A (en) * | 2016-11-14 | 2017-02-01 | 中南大学 | Strong-wind high-speed railway along-the-line wind speed space network structure prediction method |
| CN109470260A (en) * | 2018-12-13 | 2019-03-15 | 广州欧科信息技术股份有限公司 | Tour planing method, device, equipment and storage medium |
| CN112362067A (en) * | 2020-11-25 | 2021-02-12 | 江苏恒澄交科信息科技股份有限公司 | Autonomous route planning method for inland river intelligent ship |
| CN113009368A (en) * | 2021-02-24 | 2021-06-22 | 武汉科技大学 | MFF-based multi-core GPR algorithm-based lithium battery residual life prediction method |
| CN114580719A (en) * | 2022-02-23 | 2022-06-03 | 新之航传媒科技集团有限公司 | Intelligent exhibition hall interaction method and system |
-
2023
- 2023-10-30 CN CN202311429679.1A patent/CN117575118B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103376116A (en) * | 2012-04-16 | 2013-10-30 | 北京四维图新科技股份有限公司 | Scenic route planning in vehicle navigation |
| CN105046045A (en) * | 2015-06-05 | 2015-11-11 | 河海大学 | Bayesian combination-based Web service QoS prediction method |
| CN105222782A (en) * | 2015-09-07 | 2016-01-06 | 广东欧珀移动通信有限公司 | A kind of route recommendation method and mobile terminal |
| CN106372731A (en) * | 2016-11-14 | 2017-02-01 | 中南大学 | Strong-wind high-speed railway along-the-line wind speed space network structure prediction method |
| CN109470260A (en) * | 2018-12-13 | 2019-03-15 | 广州欧科信息技术股份有限公司 | Tour planing method, device, equipment and storage medium |
| CN112362067A (en) * | 2020-11-25 | 2021-02-12 | 江苏恒澄交科信息科技股份有限公司 | Autonomous route planning method for inland river intelligent ship |
| CN113009368A (en) * | 2021-02-24 | 2021-06-22 | 武汉科技大学 | MFF-based multi-core GPR algorithm-based lithium battery residual life prediction method |
| CN114580719A (en) * | 2022-02-23 | 2022-06-03 | 新之航传媒科技集团有限公司 | Intelligent exhibition hall interaction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117575118A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qin et al. | Ride-hailing order dispatching at didi via reinforcement learning | |
| CN111862579B (en) | Taxi scheduling method and system based on deep reinforcement learning | |
| Gavalas et al. | Heuristics for the time dependent team orienteering problem: Application to tourist route planning | |
| CN110555990B (en) | Effective parking space-time resource prediction method based on LSTM neural network | |
| Mokhtari et al. | Integration of efficient multi-objective ant-colony and a heuristic method to solve a novel multi-objective mixed load school bus routing model | |
| CN109117993B (en) | Processing method for optimizing vehicle path | |
| Grzybowska et al. | Decision support system for real-time urban freight management | |
| Dezfouli et al. | A novel tour planning model using big data | |
| CN109992923A (en) | A kind of transmission line of electricity paths planning method stage by stage based on variable resolution cost surface | |
| CN112418514B (en) | Method for optimizing campus bus route planning by using ant colony system | |
| CN113962472A (en) | Time-space double-attention subway passenger flow short-time prediction method based on GAT-Seq2Seq model | |
| Garcia et al. | Intelligent routing system for a personalised electronic tourist guide | |
| CN117575118B (en) | Method for planning visit route of science and technology museum | |
| CN116629413A (en) | Global optimal automatic deployment method for mobile charging device based on random forest and Dijkstra | |
| Liu et al. | A cloud-edge-end collaboration framework for cruising route recommendation of vacant taxis | |
| Qi et al. | Tourism Route Selection Model for Tourism Sustainable Development Based on Improved Genetic Algorithm | |
| Lee et al. | A hybrid system for facility layout by integrating simulation and ant colony optimization | |
| Zhou et al. | A multi-objective optimization approach to package delivery by the crowd of occupied taxis | |
| CN107563544A (en) | A kind of gray model power predicating method based on day character vector optimization | |
| Du et al. | Fuzzy bi-objective chance-constrained programming model for timetable optimization of a bus route | |
| Luo et al. | Intelligent online food delivery system: a dynamic model to generate delivery strategy and tip advice | |
| Chen et al. | [Retracted] Path Optimization Model of Rural Red Tourist Attractions Based on Ant Colony Algorithm | |
| Rashidi et al. | A hybrid ant colony system for partially dynamic vehicle routing problem | |
| Tu et al. | Deep online recommendations for connected E-taxis by coupling trajectory mining and reinforcement learning | |
| Diareme et al. | Multi-criteria optimization methods applied in agricultural touring |
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 | ||
| GR01 | Patent grant |