Route planning method, system and equipment based on guided local search algorithm and medium

Abstract

In order to solve the problems of low resource utilization efficiency and insufficient personalized service in the prior art, the invention discloses an itinerary planning method, system, equipment and medium based on a guided local search algorithm, and relates to the technical field of big data analysis and artificial intelligence, and the method comprises the steps: obtaining and preprocessing scenic spot data and traffic data; filling a playing list based on the scenic spots selected by the tourists, the scenic spot data and the traffic data; and planning a travel for the travel list based on a guided local search algorithm. Multi-source data including scenic spot opening and closing time, recommended travel duration, geographic position, scenic spot labels and the like are integrated, real-time analysis and intelligent decision are performed by applying an advanced artificial intelligence technology, high-quality travel recommendation service is provided for tourists, personalized requirements of the tourists can be met, the utilization efficiency of tourism resources can be improved, and the tourism experience of the tourists can be improved. Efficient sharing and intelligent analysis of tourism information are realized, and powerful support is provided for tourism decision making.

Description

Technical Field

[0001] The present invention relates to the field of big data analysis and artificial intelligence technology, and in particular to a trip planning method, system, device and medium based on a guided local search algorithm. Background Art

[0002] Currently, the tourism industry is developing rapidly, and many travelers face many challenges when planning their trips. They hope to fully appreciate the natural scenery, rich culture, and unique cuisine of the destination within a limited time, so careful itinerary planning is particularly important. However, itinerary planning is a complex task that requires comprehensive consideration of many factors, such as the distance between attractions, opening hours, traffic conditions, personal preferences, and weather changes. In addition, the flow of people during peak tourist periods, the carrying capacity of attractions, and some unforeseen circumstances (such as temporary closures of attractions or traffic jams) may also affect the itinerary. Therefore, a reasonable itinerary plan needs to meet individual needs while also being flexible and able to cope with emergencies.

[0003] Specifically, the following problems exist in the itinerary recommendation field: (1) Low resource utilization efficiency: The collection, processing and application methods of tourism data are relatively traditional, failing to fully tap the potential value of the data, resulting in a waste of resources. (2) Insufficient personalized services: Existing platforms and services are unable to provide itinerary recommendations that meet the personalized needs of tourists. Summary of the Invention

[0004] This invention addresses the shortcomings of the existing technology by providing an itinerary planning method, system, device, and medium based on a guided local search algorithm. This invention deeply integrates and improves tourist attraction and related service data. By utilizing advanced guided local search methods, it provides tourists with more intelligent itinerary planning services, thereby improving the convenience of travel planning and enhancing their travel experience. This invention successfully addresses the predicament of fragmented pre-trip information and complex itinerary planning, significantly enhancing tourists' overall travel experience.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for itinerary planning based on a guided local search algorithm, comprising:

[0007] Acquire and preprocess scenic spot data and traffic data;

[0008] Populate playlists based on visitor-selected attractions, attraction data, and transportation data;

[0009] Plan itineraries based on your playlist using a guided local search algorithm.

[0010] To optimize the above technical solutions, specific measures taken also include:

[0011] Furthermore, the acquisition and pre-processing of scenic spot data and traffic data is specifically as follows:

[0012] Collect scenic spot location information, opening and closing hours, recommended visiting duration, recommended visiting time interval and scenic spot labels from the online platform; crawl the latitude and longitude information of the scenic spot based on the scenic spot location information, and crawl the traffic data between two points based on the latitude and longitude information; store scenic spot location information, latitude and longitude information, opening and closing hours, recommended visiting duration, recommended visiting time interval, scenic spot labels and traffic data between scenic spots.

[0013] Furthermore, the method of filling the play list based on the tourist's selected attractions, attraction data and traffic data is as follows:

[0014] The personalized tags of tourists are obtained in combination with the attractions selected by tourists. Based on the personalized tags of tourists, the number of days selected by tourists, the location of the scenic spot and the weather conditions, the play list is filled according to the distribution of personalized tags of tourists, the distance between scenic spots and the categories of indoor and outdoor venues.

[0015] Furthermore, the itinerary planning based on the guided local search algorithm is specifically as follows:

[0016] Taking the minimum travel time as the goal, the objective function including the penalty term is established as follows:

[0017]

[0018] Where Z represents the objective function, m is the number of days of play, n is the total number of attractions visited, c ij is the commuting time from attraction i to attraction j, is a decision variable, indicating whether to go from attraction i to attraction j on day k. A value of 0 indicates not going from attraction i to attraction j, and a value of 1 indicates going from attraction i to attraction j. λ is the weight coefficient of the penalty term, which is used to balance the impact of commuting time and the penalty term. F is a feature set, where f represents the path features that need to be penalized, including pairs of attractions and congested nodes in the frequently visited path. p f is the penalty value of the path feature f, which is initially 0 and is dynamically updated with iterations. f (s) is an indicator function, which takes the value 1 if the current solution s contains the path feature f, and takes the value 0 otherwise;

[0019] The constraints of the objective function include capacity constraints, time window constraints, and scenic spot connectivity constraints;

[0020] Use a greedy algorithm to generate an initial solution that meets the constraints. According to the nearest neighbor rule, each time the nearest unvisited location is selected from the current location until the entire route is completed.

[0021] Local search: Perform preliminary optimization based on 3-opt, by deleting three edges in the path and reconnecting them, to find a solution that is better than the current path;

[0022] Update the penalty term and penalty value of the objective function;

[0023] Under the objective function with updated penalty terms and penalty values, continue local search to find a solution that is better than the current path;

[0024] Determine whether the search time limit is met. If so, the algorithm stops and obtains the final planning result. Otherwise, continue to update the penalty item and penalty value.

[0025] Furthermore, the capacity constraint is specifically that the total play time per day does not exceed the set maximum time T max , expressed as follows:

[0026]

[0027] Where, d i The duration of visiting attraction i is x ki Indicates whether attraction i is visited on day k;

[0028] The time window constraint specifically states that each attraction must be visited within its specified time window. The formula is as follows:

[0029] a i ≤t i ≤b i

[0030] Among them, a i and b i are the opening and closing times of attraction i, t i The arrival time of tourists;

[0031] The scenic spot connectivity constraint is that each scenic spot can be visited at most once, which can be expressed as follows:

[0032]

[0033] Where x ij It is a decision variable, indicating whether to go from scenic spot i to scenic spot j. A value of 0 indicates not going from scenic spot i to scenic spot j, and a value of 1 indicates going from scenic spot i to scenic spot j.

[0034] Furthermore, the penalty term and penalty value of the updated objective function are specifically:

[0035] For all features f in the current solution s, calculate the utility value using the following formula:

[0036]

[0037] Where Utility(f) is the utility value of path feature f, c f is the original cost of path feature f, the denominator is 1+p f is the normalized term of the current accumulated penalty value;

[0038] From all that I f Among the path features with (s)=1, select the path feature f with the largest utility value. * :

[0039] f * =argmax f∈F (Utility(f))

[0040] In the formula, the function argmax f∈F It means finding the value of path feature f when Utility(f) is maximized;

[0041] The path feature f with the largest utility value * , increase its penalty value:

[0042]

[0043] Where, Represents the path feature f * The updated penalty value, Represents the path feature f * The original penalty value.

[0044] The present invention also discloses a trip planning system based on a guided local search algorithm, comprising:

[0045] Information collection and maintenance module, used to obtain and pre-process scenic spot data and traffic data;

[0046] Attraction recommendation module, used to populate the playlist based on the attractions selected by tourists, attraction data and transportation data;

[0047] The itinerary generation module is used to plan itineraries for playlists based on a guided local search algorithm.

[0048] The present invention also discloses an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the itinerary planning method based on the guided local search algorithm as described above is implemented.

[0049] The present invention also discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the itinerary planning method based on the guided local search algorithm as described above.

[0050] The beneficial effects of the present invention are as follows: the present invention recommends scenic spots by deeply understanding travel preferences, scenic spot selection and other needs through scenic spots selected by users in combination with distance and weather; by integrating user personalized needs and scenic spot data, combined with a guided local search algorithm to solve the path optimization problem; the planning of the present invention takes into account constraints such as stay time and scenic spot opening hours to ensure the rationality and efficiency of the itinerary.

[0051] By combining multi-source data, this invention can provide each tourist with highly personalized itinerary recommendations, enhancing the tourist experience. It can also set time intervals to avoid peak hours based on the congestion level of scenic spots, thus realizing itinerary planning that avoids peak hours. It can also provide fine-grained travel guides based on tourists' points of interest and visiting times, thus optimizing the visiting experience. Itinerary planning algorithms can help urban traffic managers optimize traffic guidance and diversion strategies based on real-time traffic flow, road conditions, and historical data, especially during peak hours or large-scale events.

[0052] This system integrates tourism resources and plans itineraries based on tourist intent, providing highly personalized itinerary planning services and enhancing the visitor experience. This system has broad application potential, not only for optimizing tourism resources but also for urban transportation and public safety management, promoting the digital transformation and intelligent upgrading of the cultural tourism industry.

[0053] This invention not only meets the personalized needs of tourists, but also improves the utilization efficiency of tourism resources, enables efficient sharing and intelligent analysis of tourism information, and provides strong support for tourism decision-making. The implementation of this invention will significantly enhance the intelligence and personalization of itinerary planning, bring tourists a more convenient and accurate travel experience, and also inject new vitality into the high-quality development of the cultural tourism industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is an overall flow chart of the itinerary planning method based on the guided local search algorithm proposed in the present invention.

[0055] Figure 2 This is a block diagram of the itinerary planning system based on the guided local search algorithm proposed in the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0057] Example 1

[0058] The present invention proposes a method for itinerary planning based on a guided local search algorithm. The process of the method is as follows: Figure 1 As shown, including:

[0059] Acquire and preprocess scenic spot data and traffic data; specifically:

[0060] Collect scenic spot location information, opening and closing hours, recommended visiting duration, recommended visiting time interval and scenic spot labels from online platforms (such as scenic spot official website, Ctrip, Tongcheng, Xiaohongshu); crawl the latitude and longitude information of the scenic spot based on the scenic spot location information, and crawl the traffic data between two points based on the latitude and longitude information; store scenic spot location information, latitude and longitude information, opening and closing hours, recommended visiting duration, recommended visiting time interval, scenic spot labels and traffic data between scenic spots.

[0061] Populate the playlist based on the tourist's selected attractions, attraction data, and transportation data; specifically:

[0062] The personalized tags of tourists are obtained in combination with the attractions selected by tourists. Based on the personalized tags of tourists, the number of days selected by tourists, the location of the scenic spot and the weather conditions, the play list is filled according to the distribution of personalized tags of tourists, the distance between scenic spots and the categories of indoor and outdoor venues.

[0063] Plan itineraries based on guided local search algorithms. Specifically:

[0064] Taking the minimum travel time as the goal, the objective function including the penalty term is established as follows:

[0065]

[0066] Guided local search modifies the objective function by introducing a penalty term, so that the search process avoids premature convergence to the local optimal solution. The new objective function will guide the local search away from frequently selected features, increase diversity, and thus explore more potential solutions. Where Z represents the objective function, m is the number of days to play, n is the total number of attractions to play, and c is the number of times to play. ij is the commuting time from attraction i to attraction j, is a decision variable, indicating whether to go from attraction i to attraction j on day k. A value of 0 indicates not going from attraction i to attraction j, and a value of 1 indicates going from attraction i to attraction j. λ is the weight coefficient of the penalty term, which is used to balance the impact of commuting time and the penalty term. F is a feature set, where f represents the path features that need to be penalized, including pairs of attractions and congested nodes in the frequently visited path. By increasing the penalties for these features, the GLS guides the algorithm to avoid over-reliance on these features, thereby exploring new solution spaces. f is the penalty value of the path feature f, which is initially 0 and is dynamically updated with iterations. f (s) is an indicator function, which takes the value 1 if the current solution s contains the path feature f, and takes the value 0 otherwise;

[0067] The constraints of the objective function include capacity constraints, time window constraints, and scenic spot connectivity constraints. The capacity constraint specifically states that the total daily play time does not exceed the set maximum time T max , expressed as follows:

[0068]

[0069] Where, d i The duration of visiting attraction i is x ki Indicates whether attraction i is visited on day k.

[0070] The time window constraint specifically means that each attraction must be visited within its specified time window. The formula is as follows:

[0071] a i ≤t i ≤b i

[0072] Among them, a i and b i are the opening and closing times of attraction i, t i Arrival time for tourists.

[0073] The specific scenic spot connectivity constraint is that each scenic spot must be visited at most once, which can be expressed as follows:

[0074]

[0075] Where x ij It is a decision variable, indicating whether to go from scenic spot i to scenic spot j. A value of 0 indicates not going from scenic spot i to scenic spot j, and a value of 1 indicates going from scenic spot i to scenic spot j.

[0076] Use a greedy algorithm to generate an initial solution that satisfies the constraints. Based on the nearest neighbor rule, each time the closest unvisited location is selected from the current location until the entire route is completed. This solution may not be the global optimal solution, but it can serve as a good foundation for subsequent optimization.

[0077] Local search: Performs preliminary optimization based on 3-opt, searching for a solution that is better than the current path by deleting three edges in the path and reconnecting them. Its essence is to reduce the total path time by path reconstruction.

[0078] Update the penalty term and penalty value of the objective function; the core idea of ​​GLS is to solve the local optimal problem by introducing a penalty mechanism to prevent the algorithm from falling into the local optimal state. Specifically:

[0079] For all features f in the current solution s, calculate the utility value using the following formula:

[0080]

[0081] Where Utility(f) is the utility value of path feature f, c f is the original cost of the path feature f (such as commuting time or congestion). The higher the cost, the more penalty is required. The denominator is 1+p f It is the normalization term of the current cumulative penalty value. The larger the penalty value, the lower the utility (to avoid over-punishing the same feature).

[0082] From all that I f Among the path features where (s) = 1 (i.e., the current solution contains path feature f), select the path feature f with the largest utility value. * :

[0083] f * =argmax f∈F (Utility(f))

[0084] In the formula, the function argmax f∈F It means finding the value of the path feature f when Utility(f) is maximized. The significance of this step is to give priority to punishing the features that have the greatest impact on the quality of the current solution and have not yet been fully punished.

[0085] The path feature f with the largest utility value * , increase its penalty value:

[0086]

[0087] Where, Represents the path feature f * The updated penalty value, Represents the path feature f * The original penalty value.

[0088] Under the updated objective function with the penalty term and penalty value, the local search continues to find a solution that is better than the current path. Guided by the new objective function, the local search algorithm prioritizes unpenalized solutions, thus avoiding local optimality. Simultaneously, as the objective function is continuously revised, the search process gradually converges to the global optimal solution.

[0089] Determine whether the search time limit is met. If so, the algorithm stops and obtains the final planning result. Otherwise, continue to update the penalty item and penalty value.

[0090] Unlike traditional itinerary planning algorithms, this method considers more planning factors. The contributions of this method are:

[0091] (1) Consider the actual opening and closing times of the scenic spot, the recommended duration of the visit, the recommended visiting range, and other factors;

[0092] (2) Taking into account the real traffic conditions;

[0093] (3) Based on the scenic spots selected by tourists, intelligently recommend and fill in itineraries to enrich the itinerary content.

[0094] Example 2

[0095] The present invention proposes a trip planning system based on a guided local search algorithm corresponding to the method of embodiment 1. The system block diagram is as follows: Figure 2 As shown, including:

[0096] Information collection and maintenance module, used to obtain and pre-process scenic spot data and traffic data;

[0097] Attraction recommendation module, used to populate the playlist based on the attractions selected by tourists, attraction data and transportation data;

[0098] The itinerary generation module is used to plan itineraries for playlists based on a guided local search algorithm.

[0099] The implementation of each module and module function in the system is completely consistent with the steps of the method in Example 1, so it will not be repeated here.

[0100] Example 3

[0101] The present invention proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the itinerary planning method based on the guided local search algorithm as described in the first embodiment is implemented.

[0102] Example 4

[0103] The present invention provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the itinerary planning method based on the guided local search algorithm as described in the first embodiment.

[0104] In the embodiments disclosed herein, computer storage media can be tangible media that can contain or store programs for use by or in conjunction with an instruction execution system, device, or apparatus. Computer storage media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. More specific examples of computer storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0105] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0106] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A trip planning method based on a guided local search algorithm, characterized in that: include: Acquire and preprocess scenic spot data and traffic data; Populate playlists based on visitor-selected attractions, attraction data, and transportation data; Plan itineraries based on your playlist using a guided local search algorithm.

2. The itinerary planning method based on the guided local search algorithm according to claim 1, characterized in that: The acquisition and pre-processing of scenic spot data and traffic data are specifically as follows: Collect scenic spot location information, opening and closing hours, recommended visiting duration, recommended visiting time interval and scenic spot labels from the online platform; crawl the latitude and longitude information of the scenic spot based on the scenic spot location information, and crawl the traffic data between two points based on the latitude and longitude information; store scenic spot location information, latitude and longitude information, opening and closing hours, recommended visiting duration, recommended visiting time interval, scenic spot labels and traffic data between scenic spots.

3. The itinerary planning method based on the guided local search algorithm according to claim 1, characterized in that: The method of filling the play list based on the tourist's selected attractions, attraction data and transportation data is as follows: The personalized tags of tourists are obtained in combination with the attractions selected by tourists. Based on the personalized tags of tourists, the number of days selected by tourists, the location of the scenic spot and the weather conditions, the play list is filled according to the distribution of personalized tags of tourists, the distance between scenic spots and the categories of indoor and outdoor venues.

4. The itinerary planning method based on the guided local search algorithm according to claim 1, wherein: The specific steps of planning the playlist based on the guided local search algorithm are as follows: Taking the minimum travel time as the goal, the objective function including the penalty term is established as follows: Where Z represents the objective function, m is the number of days of play, n is the total number of attractions visited, c ij is the commuting time from attraction i to attraction j, is a decision variable, indicating whether to go from attraction i to attraction j on day k. A value of 0 indicates not going from attraction i to attraction j, and a value of 1 indicates going from attraction i to attraction j. λ is the weight coefficient of the penalty term, which is used to balance the impact of commuting time and the penalty term. F is a feature set, where f represents the path features that need to be penalized, including pairs of attractions and congested nodes in the frequently visited path. p f is the penalty value of the path feature f, which is initially 0 and is dynamically updated with iterations. f (s) is an indicator function, which takes the value 1 if the current solution s contains the path feature f, and takes the value 0 otherwise; The constraints of the objective function include capacity constraints, time window constraints, and scenic spot connectivity constraints; Use a greedy algorithm to generate an initial solution that meets the constraints. According to the nearest neighbor rule, each time the nearest unvisited location is selected from the current location until the entire route is completed. Local search: Perform preliminary optimization based on 3-opt, by deleting three edges in the path and reconnecting them, to find a solution that is better than the current path; Update the penalty term and penalty value of the objective function; Under the objective function with updated penalty terms and penalty values, continue local search to find a solution that is better than the current path; Determine whether the search time limit is met. If so, the algorithm stops and obtains the final planning result. Otherwise, continue to update the penalty item and penalty value.

5. The itinerary planning method based on the guided local search algorithm according to claim 4, characterized in that: The capacity constraint is specifically that the total play time per day does not exceed the set maximum time T max , expressed as follows: Where, d i The duration of visiting attraction i is x ki Indicates whether attraction i is visited on day k; The time window constraint specifically states that each attraction must be visited within its specified time window. The formula is as follows: a i ≤t i ≤b i Among them, a i and b i are the opening and closing times of attraction i, t i The arrival time of tourists; The scenic spot connectivity constraint is that each scenic spot can be visited at most once, which can be expressed as follows: Where x ij It is a decision variable, indicating whether to go from scenic spot i to scenic spot j. A value of 0 indicates not going from scenic spot i to scenic spot j, and a value of 1 indicates going from scenic spot i to scenic spot j.

6. The itinerary planning method based on the guided local search algorithm according to claim 4, characterized in that: The penalty term and penalty value of the updated objective function are specifically: For all features f in the current solution s, calculate the utility value using the following formula: Where Utility(f) is the utility value of path feature f, c f is the original cost of path feature f, the denominator is 1+p f is the normalized term of the current accumulated penalty value; From all that I f Among the path features with (s)=1, select the path feature f with the largest utility value. * : f * =argmax f∈F (Utility(f)) In the formula, the function argmax f∈F It means finding the value of path feature f when Utility(f) is maximized; The path feature f with the largest utility value * , increase its penalty value: Where, Represents the path feature f * The updated penalty value, Represents the path feature f * The original penalty value.

7. A trip planning system based on a guided local search algorithm, characterized in that: include: Information collection and maintenance module, used to obtain and pre-process scenic spot data and traffic data; Attraction recommendation module, used to populate the playlist based on the attractions selected by tourists, attraction data and transportation data; The itinerary generation module is used to plan itineraries for playlists based on a guided local search algorithm.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the itinerary planning method based on the guided local search algorithm according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: The computer program enables a computer to execute the itinerary planning method based on the guided local search algorithm according to any one of claims 1 to 6.

Images