A method and system for land space planning management based on big data

Abstract

This disclosure provides a method and system for land spatial planning management based on big data. The method includes: collecting historical traffic flow data and land use evolution data, and preprocessing and standardizing them; constructing a spatiotemporal evolution model based on the historical traffic flow and land use evolution trends, and identifying potential traffic growth hotspots; performing time-series regulation of transportation facility corridor space based on the prediction results of the spatiotemporal evolution model and generating dynamic reservation schemes; and feeding the dynamic reservation schemes back to the land spatial planning system to guide the timing of infrastructure construction. Through the solution of this disclosure, the time-series regulation of transportation facility corridor space reservation can be achieved based on historical traffic flow and land use evolution trends to solve the problems of premature infrastructure construction and resource waste.

Description

Technical Field

[0001] This application relates to the field of land and resources management technology, specifically to a land and space planning management method and system based on big data. Background Technology

[0002] In traditional land spatial planning and management, the spatial reservation of transportation facility corridors relies heavily on static planning or expert experience, lacking a coordinated analysis of the dynamic changes in historical traffic flow and the evolution trend of land use. This often leads to a disconnect between reservation plans and actual needs, resulting in the premature construction of infrastructure leading to idle and wasted resources, or insufficient reservations that cannot meet future traffic growth demands. Existing technologies struggle to accurately identify potential traffic growth hotspots and to scientifically regulate the timing of corridor reservations, making the planning system inflexible and lacking foresight in responding to dynamic urban development, and unable to achieve a precise match between the timing of infrastructure construction and the pace of regional development. Summary of the Invention

[0003] In view of this, the present disclosure provides a land spatial planning management method and system based on big data, which at least partially solves the problems existing in the prior art.

[0004] The first invention discloses a land spatial planning management method based on big data, comprising:

[0005] Historical traffic flow data and land use evolution data are collected, and preprocessed and standardized.

[0006] Based on the historical traffic flow and land use evolution trends, a spatiotemporal evolution model is constructed and potential traffic growth hotspots are identified.

[0007] Based on the prediction results of the spatiotemporal evolution model, the space of transportation facility corridors is subject to temporal regulation and a dynamic reservation scheme is generated.

[0008] The dynamic reservation scheme will be fed back into the land and space planning system to guide the timing of infrastructure construction.

[0009] According to one embodiment, constructing a spatiotemporal evolution model based on historical traffic flow and land use evolution trends includes:

[0010] Obtain the traffic flow data matrix D for each region within different time periods;

[0011] Obtain the land use type distribution matrix U for the corresponding time period;

[0012] Feature extraction is performed on the data using a sliding time window, and the time-series correlation coefficient matrix is ​​calculated. Where N is the number of sampling points;

[0013] Determine whether a land use structure change has occurred based on a threshold Δ: If... Then proceed to the next stage.

[0014] According to one embodiment, the temporal regulation of transportation facility corridor space based on the prediction results of the spatiotemporal evolution model includes:

[0015] Calculate the current regional traffic load rate ρ = actual traffic flow / expected maximum traffic volume;

[0016] Constructing the time decay factor , where t is the step size from the current time, and τ is the preset adjustment time constant;

[0017] Calculate the reserved space weights based on ρ and λ. ;

[0018] When W is greater than the set threshold η, it is marked as a priority allocation area; otherwise, allocation is postponed.

[0019] According to one embodiment, feeding back the dynamic reservation scheme to the land and space planning system further includes:

[0020] Generate a traffic corridor reservation level sequence Each level corresponds to a different construction time window length;

[0021] Using a weighted entropy model The scheme is optimized and selected, where p_i represents the probability of a certain level being selected;

[0022] The final dynamic scheme to be retained is determined based on the size of S. The scheme corresponding to the smallest S is the optimal one. If the actual construction progress deviates from the scheme by more than ε, the readjustment process is initiated.

[0023] According to one embodiment, the temporal regulation of the transportation facility corridor space further includes:

[0024] Extracting historical traffic flow growth rate ;

[0025] Set an upper limit for land carrying capacity , where A is the area of ​​the region and α is the land carrying capacity coefficient;

[0026] Assess the congestion risk in the current area When M > β, the dynamic reservation procedure is activated, where β is the early warning threshold.

[0027] The proportion of reserved resources to be allocated is determined by the ratio of M to K: , where δ is the control coefficient.

[0028] According to one embodiment, the generation of the dynamic reservation scheme further includes:

[0029] Establish a traffic demand growth model , where T_b is the time period number;

[0030] Based on the degree of land use change Assess the activity level of spatial changes;

[0031] The model output Y is input into the dynamic allocation function. , where γ is the adjustment parameter;

[0032] Traffic demand is categorized into different levels based on the Q value, with higher-level demands given priority in response.

[0033] According to one embodiment, the temporal regulation of the transportation facility corridor space further includes:

[0034] Analysis of road utilization rate indicators Where v_i is the number of vehicles within a certain period of time, and v_ideal is the ideal traffic volume;

[0035] Establish resource allocation equations , where L_w is the width, L_a is the average density, and t_w is the time weight;

[0036] Resource gaps are assessed based on R_d and prioritized.

[0037] If R_d > 1, it means that expansion is needed; otherwise, there is no need to intervene immediately.

[0038] According to one embodiment, the construction of the spatiotemporal evolution model further includes:

[0039] Collect spatial information from multiple dimensions, including road network structure S_r, land use type U_l, and the proportion of different modes of transportation M_t;

[0040] Calculate the similarity matrix between each dimension ;

[0041] Construct a weight matrix by incorporating time-varying factors , where δ controls the rate of change, and t_0 is the starting time;

[0042] By combining similarity with time weights, a multi-dimensional fusion feature matrix is ​​formed. It is used for modeling and analysis.

[0043] The second invention discloses a land spatial planning management system based on big data, comprising:

[0044] The data acquisition module collects historical traffic flow data and land use evolution data, and performs preprocessing and standardization.

[0045] The model building module constructs a spatiotemporal evolution model based on the historical traffic flow and land use evolution trends, and identifies potential traffic growth hotspots.

[0046] The control module performs temporal control on the transportation facility corridor space based on the prediction results of the spatiotemporal evolution model and generates a dynamic reservation scheme.

[0047] The feedback module feeds the dynamic reservation scheme back to the land and space planning system to guide the timing of infrastructure construction.

[0048] This disclosure provides a big data-based land spatial planning and management method, including: collecting historical traffic flow data and land use evolution data, and preprocessing and standardizing them; constructing a spatiotemporal evolution model based on the historical traffic flow and land use evolution trends, and identifying potential traffic growth hotspots; performing time-series regulation of transportation facility corridor space based on the prediction results of the spatiotemporal evolution model and generating dynamic reservation schemes; and feeding the dynamic reservation schemes back to the land spatial planning system to guide the timing of infrastructure construction. Through the solution of this disclosure, the time-series regulation of transportation facility corridor space reservation can be achieved based on historical traffic flow and land use evolution trends to address the problems of premature infrastructure construction and resource waste. Attached Figure Description

[0049] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments disclosed in this application and should not be construed as limiting the scope of this application.

[0050] Figure 1 This is a flowchart of a land spatial planning management method based on big data according to the present invention;

[0051] Figure 2 This is a flowchart of the spatiotemporal evolution model constructed based on historical traffic flow and land use evolution trends in this invention;

[0052] Figure 3 This is a flowchart of the present invention for temporal regulation of transportation facility corridor space based on the prediction results of the spatiotemporal evolution model;

[0053] Figure 4 This is a flowchart of the present invention that feeds back the dynamic reservation scheme to the land and space planning system;

[0054] Figure 5 This is a flowchart of the present invention for the temporal regulation of transportation facility corridor space;

[0055] Figure 6This is a flowchart of the dynamic reservation scheme generation method of the present invention;

[0056] Figure 7 This is a flowchart of the present invention for the temporal regulation of transportation facility corridor space;

[0057] Figure 8 This is a flowchart of the spatiotemporal evolution model constructed according to the present invention;

[0058] Figure 9 This is a block diagram of the land spatial planning management system based on big data of the present invention. Detailed Implementation

[0059] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0060] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0061] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0062] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The drawings only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0063] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0064] Next, refer to Figure 1 This invention describes a land spatial planning management method based on big data, which includes the following steps:

[0065] S101: Collect historical traffic flow data and land use evolution data, and preprocess and standardize them. This process involves collecting road traffic data over many years, traffic hub flow data, and records of various urban land use changes to ensure data comprehensiveness and timeliness. The collected data is then cleaned to remove invalid or erroneous data, such as outliers, duplicate data, and missing fields, and standardized according to a unified format and time unit to provide a consistent foundation for subsequent analysis. For example, multi-source heterogeneous data can be integrated through a GIS (Geographic Information System) platform and converted into a standard input format suitable for machine learning models.

[0066] Specifically, historical traffic flow data mainly comes from fixed-point detectors (such as loop detectors and microwave radars) deployed by urban traffic management departments, GPS trajectory data from floating vehicles, mobile communication signaling data, and records from major traffic checkpoints, to cover the complete urban development cycle. Land use evolution data is mainly obtained from the annual national land survey database, remote sensing image interpretation results, and urban planning archives of the natural resources department. Data types include, but are not limited to, the distribution and area changes of different land uses such as residential, commercial, industrial, and public facilities. After data collection, the data enters the preprocessing and standardization stage: First, data cleaning is performed to remove outliers caused by equipment failure (such as instantaneous traffic flow exceeding the theoretical capacity of the road) and to fill in short-term data gaps caused by communication interruptions (using time series interpolation). Then, spatial alignment is performed to unify all data to the same geographic coordinate system (such as the CGCS2000 National Geodetic Coordinate System) and the same spatial grid unit (such as a regular grid of 500m×500m). Finally, time alignment and standardization are performed to unify data with different collection frequencies (such as traffic flow per minute and land use per year) to the same analysis time granularity (such as quarterly or annual) through resampling and aggregation, and to normalize numerical data to eliminate the influence of units.

[0067] S102: Construct a spatiotemporal evolution model based on the historical traffic flow and land use evolution trends, and identify potential traffic growth hotspots.

[0068] Specifically, firstly, using spatiotemporal series analysis methods, time series of traffic flow and time series of land use area proportions are established for each spatial grid unit. Next, methods such as Granger causality tests are used to quantitatively analyze whether land use change is a leading factor in traffic flow change. Then, an integrated model combining spatiotemporal autoregression and cross-correlation, such as a "spatiotemporal weighted regression model," is constructed. This model considers not only the historical traffic flow of a specific location but also the traffic flow of surrounding adjacent areas during the same period and the current and historical land use structures as explanatory variables. By training this model, the response coefficient of traffic flow to land use change and its spatial heterogeneity can be obtained. After the model is run, its predictive function is used to simulate the spatial distribution of traffic demand over the next 5-15 years under different future scenarios (such as the establishment of a planned commercial center or the construction of a large residential area). By comparing the difference between future demand and current supply, and identifying continuous spatial clusters where the demand growth rate is consistently higher than the average level, these can be identified as "potential traffic growth hotspots."

[0069] Taking the development of a specific new urban area as an example, the process begins by collecting traffic flow data generated by traffic monitors and floating car GPS for each quarter over the past ten years, as well as land use data (such as changes in commercial, residential, and industrial land areas) from historical land surveys and remote sensing images. This data is then uniformly gridded into 500m × 500m units, forming a traffic flow time-series matrix D and a land use type distribution matrix U. Next, using a sliding time window with a three-year step size, the time-series correlation coefficient matrix C between D and U is calculated window by window to quantify the correlation strength between traffic and land use evolution. Simultaneously, multi-dimensional spatial information such as road network density, bus stop distribution, and the proportion of different modes of transportation are introduced. The similarity matrix M_sim between each dimension is calculated using cosine similarity and then integrated. A multidimensional fusion feature matrix A is generated by applying time decay weights W_t (giving higher weights to recent data). Based on matrix A, a spatiotemporal weighted regression model is constructed to train historical data to fit the spatial heterogeneity response pattern of traffic flow with land use changes, and to simulate the distribution of traffic demand under the scenario of the planned commercial center taking place in the next five years. Finally, based on the grid-level traffic growth rate prediction value output by the model, an annual average growth rate threshold Δ=15% is set, and a spatial clustering algorithm is used to identify contiguous areas where the growth rate of three or more consecutive grid units exceeds the threshold and land use changes are active (e.g., the correlation coefficient C is consistently higher than the historical average), such as the new city center planned in the eastern part of the city. These areas are marked as potential traffic growth hotspots, providing precise targets for subsequent corridor reservations.

[0070] S103: Based on the prediction results of the spatiotemporal evolution model, the space of the transportation facility corridor is subject to temporal regulation and a dynamic reservation scheme is generated.

[0071] Specifically, first, the current traffic load rate (ρ) of a specific area is calculated, which is the ratio of the current actual traffic flow to the road's designed capacity, to assess its current urgency. Second, based on the model's predicted future traffic growth curve, a "demand urgency point" is defined—the year in which the predicted traffic flow will reach the design capacity threshold. Next, the concepts of a time decay factor (λ) and reserved space weight (W) are introduced. For example, for a corridor predicted to reach saturation in 10 years, its current construction urgency should be lower than that of a corridor that will reach saturation in 3 years. During implementation, a comprehensive priority score is calculated for each corridor plan, which integrates multiple weighting factors such as the current load rate, the remaining time to reach saturation, and the importance of the land use function served by the corridor (e.g., whether it connects to emergency facilities). Based on this score, the system automatically generates a "dynamic reservation plan." This plan is not a static "red line," but a "plan package" containing different implementation sequences. For example, for the highest priority corridors, the plan requires the immediate delineation of strictly controlled construction reserve space in the national land space planning, prohibiting other uses from occupying it. For medium priority corridors, flexible control zones can be delineated, allowing temporary low-intensity use in the near term, but preserving conditions for long-term renovation. For lower priority corridors, only strategic marking is done, and they will be evaluated again after the next round of model updates.

[0072] Taking a specific urban planning case as an example, after predicting that the traffic demand in the eastern new area of ​​the city will increase by 80% in the next five years based on a spatiotemporal evolution model, the specific process of implementing time-series regulation and generating dynamic reserve schemes is as follows: First, for the three candidate traffic corridors (A, B, and C) connecting the new area, their current traffic load rate ρ is calculated (e.g., corridor A ρ=0.8, close to saturation; corridor B ρ=0.4, relatively idle; corridor C ρ=0.6, moderate pressure), and combined with the "demand urgency time point" predicted by the model for each corridor to reach the design capacity threshold (e.g., A in 2 years, B in 8 years, and C in 5 years); then, a time decay factor is constructed. Where t is the time remaining until the urgency point, and τ is set to 5 years (reflecting the forward-looking nature of the plan), the calculated λ value is smaller for corridor A (high urgency), medium for corridor C, and larger for corridor B (low urgency). Subsequently, considering the load factor and time decay factor, the λ value is calculated using the formula... Calculate the reserved space weight W for each corridor (where 0.5 is the benchmark load rate), obtaining WA = 0.72 for Corridor A, WC = 0.58 for Corridor C, and WB = 0.31 for Corridor B; when setting the threshold η = 0.5, Corridors A and C are marked for priority allocation, and Corridor B is postponed; based on this weight ranking, the system automatically generates a dynamic reservation plan: for Corridor A (W > 0.7), the plan demarcates a strict construction control zone with a width of 60 meters, requiring immediate inclusion in the core control area of the national land space planning and prohibiting other projects from occupying it, and it is recommended to start the construction of the expressway within 2 years; for Corridor C (0.5 < W < 0.7), a 40-meter flexible control zone is demarcated, allowing it to be used as a temporary green space or for low-intensity development in the near future, but requiring the retention of the underground space ownership and starting construction after a review based on the actual traffic flow within 5 years; for Corridor B (W < 0.5), only the line position of a 20-meter strategic reserved passage is marked in the plan, and a detailed assessment will be carried out after 8 years. This plan also sets the weight W to be updated annually in combination with the latest traffic data. If the traffic growth rate in the area where Corridor B is located exceeds expectations due to a sudden large project, its priority will be automatically increased and the reservation plan will be adjusted through a feedback mechanism, so as to achieve precise and dynamic timing control of the infrastructure corridor space.

[0073] S104: Feed back the dynamic reservation plan to the national land space planning system to guide the timing arrangement of infrastructure construction. This part involves the integration and coordination of the system. The prediction results are directly uploaded to the national land space basic information platform as the technical basis for the next approval process. This enables government departments to grasp the future development direction at an early stage, thus more scientifically and reasonably allocate financial budgets and land indicators. For example, in a new city construction project, according to the prediction, the core traffic artery should not need to start formal construction until five years later. Therefore, the relevant sections can be included in the reserved buffer management scope in the early stage to prevent being occupied by temporary development, and at the same time guide the direction of real estate development to achieve efficient resource allocation. This method can also enhance the resilience and sustainable development ability of the city, and reduce the environmental risks and social costs brought by unordered expansion.

[0074] The beneficial effects of this technical solution are as follows. Through in-depth analysis and application of big data, it has achieved precise grasp of the layout of transportation facilities and land use changes, solved the problem of resource waste caused by information lag or one-sided decision-making in traditional planning, significantly improved the foresight, scientificity and feasibility of urban planning, and provided strong support for realizing high-quality urban construction.

[0075] Next, refer to Figure 2 , and describe the steps of constructing a spatio-temporal evolution model according to historical traffic flow and land use evolution trends of the present invention.

[0076] S201: Obtain the traffic flow data matrix D for each area within different time periods. This matrix records the traffic flow in each area at different times and is typically collected by road network monitoring equipment, taxi GPS, mobile network signals, etc. For example, in an urban planning case, the vehicle traffic volume during the morning rush hour in a certain city over the past five years can be collected as data input.

[0077] S2021: Obtain the land use type distribution matrix U for the corresponding time period. This matrix records the main land use types of each plot during the same period, such as commercial, residential, industrial, and green space, and is usually obtained from government or GIS databases. For example, in one embodiment, land use changes in each area can be obtained through remote sensing image analysis, providing basic information for subsequent modeling.

[0078] S203: Extract features from the data using a sliding time window and calculate the time-series correlation coefficient matrix. ,in This is the transpose of the traffic flow matrix, where U is the land use matrix and N is the number of sampling points. This formula quantifies the relationship between traffic flow and land use, with parameter N controlling the degree of averaging in the results. The maximum value in the formula represents the strongest correlation between the two, and a suitable threshold Δ is set to determine whether structural changes have occurred.

[0079] S204: Determine whether a change in land use structure has occurred based on a threshold Δ. If max(C) > Δ, it indicates that the traffic and land use patterns in the area have changed significantly, requiring subsequent planning adjustments. For example, in a specific embodiment, if the traffic flow on the main road in a development zone increases significantly and is accompanied by an increase in the proportion of commercial land, the system determines that the area's function has transformed, and an updated planning scheme should be considered.

[0080] In other words, in this application, the preprocessed data is first divided into multiple time periods at fixed intervals (e.g., annually). For each time period, two key matrices are constructed: a traffic flow data matrix D, where rows represent different observation points (e.g., traffic monitoring sections) and columns represent traffic flow values ​​at different sampling times within that time period (e.g., daily peak hours); and a land use type distribution matrix U, where rows are spatially aligned with matrix D (same observation area) and columns represent the area proportion of different land use types (e.g., commercial, residential, industrial). Next, a sliding time window with a length of three time periods is set and slid sequentially along the time axis. Within each window, the temporal correlation coefficient matrix C between the traffic matrix D and the land use matrix U is calculated using the following formula: ,in Let C be the transpose of D, and N be the total number of sampling points within the window. This operation quantifies the intensity and pattern of the interaction between traffic flow and various land uses within a specific time period. Then, a judgment threshold Δ (e.g., Δ=0.35) is set based on historical data analysis. After calculation for each time window, the maximum value in matrix C is checked: if max(C)>Δ, it is determined that a significant change in land use structure has occurred in the area within this time window, and this signal will trigger the system to enter the next stage of in-depth evolutionary modeling.

[0081] This technical solution combines multi-source data to achieve real-time monitoring and intelligent identification of the evolution of national land space, improving the scientific nature and foresight of land planning and facilitating precise policy implementation for optimal resource allocation and urban development management.

[0082] Next, refer to Figure 3 This invention describes a scheme for temporal regulation of transportation facility corridor space based on the prediction results of a spatiotemporal evolution model.

[0083] S301: Calculate the current regional traffic load rate ρ = actual traffic flow / expected maximum capacity. Here, actual traffic flow is the real-time number of vehicles within a certain time period, and expected maximum capacity is the maximum capacity of that road segment. ρ typically ranges between 0 and 1; when ρ approaches 1, it indicates that traffic is nearing saturation. This parameter is used to assess the current traffic pressure state; if the load rate is too high, it will affect traffic flow. For example, during peak hours on elevated roads in a city, data is collected through an intelligent monitoring system to calculate the ρ value for each road segment in real time.

[0084] S302: Constructing the time decay factor λ is the time step from the current point in hours, and τ is a preset adjustment time constant, typically ranging from 2 to 4 hours. As t increases, λ decreases, indicating that the longer the time, the lower the demand for regulation. This formula is used to simulate regulation preferences that change over time. For example, if a major event causes short-term congestion, as traffic recovers over time, the demand for emergency regulation gradually weakens, and the value of λ decreases accordingly.

[0085] S303: Calculate the reserved space weights based on ρ and λ When W is greater than the set threshold η, it is marked as a priority allocation area; otherwise, allocation is delayed. Here, (1-λ) reflects the effect of time decay, while λ reflects the basic contribution of the time factor. The higher W is, the more priority resources need to be allocated to that area. For example, during peak traffic periods on holidays, some road sections may experience a sudden surge in traffic flow, causing W to exceed the set threshold η (e.g., 0.55). These sections are marked as priority allocation areas to divert traffic in advance and avoid complete paralysis.

[0086] In other words, in this embodiment, time-series control calculations are performed for each traffic facility corridor (or specific road segment) to be evaluated. First, the current traffic load rate ρ = Q_actual / Q_max is calculated, where Q_actual is the actual hourly traffic flow obtained through monitoring, and Q_max is the expected maximum capacity of the road after its design or current modifications. This indicator directly reflects the urgency of the current situation. Second, a time decay factor is constructed. Here, t represents the time step (in years) from the current moment until the model predicts that the corridor's traffic demand will reach saturation capacity. τ is a preset adjustment time constant (e.g., τ = 5 years), which controls the rate at which the urgency of future demand decays over time. The closer λ is to 1, the more "distant" the future impact, and the lower the urgency of current regulation. Subsequently, the comprehensive weight of the corridor's reserved space is calculated based on ρ and λ. In the formula, (ρ-0.5) centers the current load rate at a baseline of 0.5 and then multiplies it by (1-λ), giving higher weighting to corridors with high current pressure and imminent future urgency. Finally, a control threshold η is set (e.g., η=0.55). The calculated W value is compared with η: if W>η, the area where the corridor is located is marked as a "priority allocation area," meaning that space resources need to be reserved or guaranteed for it in the near-term planning; otherwise, it is marked as a "delayed allocation area," included in the long-term or flexible reservation considerations. By introducing the load rate ρ and the time decay factor λ, the formula creatively integrates the current pressure and the timing of future demand, and achieves scientific ranking through a comprehensive weight W. This avoids the one-sidedness of making decisions based solely on current congestion or solely on long-term forecasts, making the priority determination of corridor space allocation reflect both current urgency and long-term needs, achieving precise optimization of resource allocation in the time dimension.

[0087] Next, refer to Figure 4 This invention describes a scheme for feeding back dynamic reservation schemes into the land and space planning system.

[0088] S401: Generate a traffic corridor reservation level sequence Each level corresponds to a different construction time window length. This sequence divides reserved land parcels into hierarchical categories by analyzing factors such as traffic demand, population distribution, and land use change. Parameters arrive The value represents different reservation priorities; a larger value indicates a higher priority for construction. Each level P_i corresponds to a specific recommended construction time window length (e.g., ...). Corresponding to 1-3 years, Corresponding to 4-7 years, (Corresponding to 8 years or more). For example, in one embodiment, P can include three levels: near-term, medium-term, and long-term, with corresponding construction time windows of 3 years, 5 years, and 10 years, respectively.

[0089] S402: Employing a weighted entropy model Optimize and select the appropriate solution, among which This indicates the probability of a certain level being selected; more specifically... This indicates a specific reservation level under a particular scheme. The probability of the total amount of resources allocated relative to the total expenditure (which can be obtained by normalizing the proportion of investment estimates at each level). The value range of is [0,1], and The minimum value of S corresponds to the optimal choice because it reflects the uniformity of the probability distribution.

[0090] In one specific embodiment, it is assumed that three alternative dynamic reservation schemes (Scheme A, Scheme B, and Scheme C) are generated for optimization selection. Each scheme classifies the same set of traffic corridors (e.g., into emergency and emergency). ,conventional Postponement (Three levels) and formulated corresponding investment or spatial resource allocation plans.

[0091] First, for each option, estimate or plan the allocation to each reservation level. The total amount of resources. Resources can be expressed as total investment budget, reserved land area, or standardized "construction equivalent". For example:

[0092] Plan A: Allocate 50% of the budget to (Urgent), 30% given (Regular) 20% given (Postponed).

[0093] Plan B: Allocate 30% of the budget to 40% given 30% given .

[0094] Plan C: Allocate 70% of the budget to 20% given 10% .

[0095] Next, for any given scheme, the resource allocation to each level is normalized to ensure that the sum of the allocation ratios for all levels is 1. This normalized ratio is the resource allocation for the corresponding level under that scheme. .

[0096] Taking Option A as an example:

[0097] ( The resource allocation ratio for each level is 0.50.

[0098] ( The percentage of resources allocated to each level is 0.30.

[0099] ( The percentage of resources allocated to each level is 0.20.

[0100] Specifically, in a big data-based transportation planning scenario, by analyzing the actual development data of different regions and calculating the selection probability of each level, the optimal reservation scheme with the lowest entropy value can be obtained.

[0101] S403: The final dynamic scheme to be retained is determined based on the value of S. The scheme with the smallest S is the optimal one, and if the deviation between the actual construction progress and the scheme is greater than ε, the readjustment process is initiated. If the value of S is low, it indicates that the scheme is more balanced and reasonable in terms of risk control and resource allocation. For example, in one embodiment, if the value of S is 0.85, the scheme is considered to be the most feasible at present and is included in the land and space planning system.

[0102] Finally, set the feedback trigger condition: if the actual construction progress deviates from the plan by more than ε, then the readjustment process will be initiated. ε is a set threshold used to measure the degree to which the actual construction situation deviates from the plan. For example, if ε is set to 5%, then when the actual progress of the construction project deviates from the reserved plan by more than 5%, a new evaluation and plan adjustment process will be initiated.

[0103] This technical solution can improve the dynamic adaptability of land and space planning, reduce errors caused by information lag or changes, enhance the flexibility and efficiency of resource allocation, and improve the scientificity and rationality of planning and management.

[0104] Next, wipe it off. Figure 5 This describes the steps of the present invention for temporal regulation of transportation facility corridor space.

[0105] S501: Extract historical traffic flow growth rate Where D_t represents traffic flow data at a specific point in time, and r represents the increase in traffic volume per unit time. This parameter generally ranges from 0 to 1, with the optimal value adjusted based on actual conditions. Typically, a lower value is set within a reasonable growth range to avoid excessive fluctuations. The formula measures the growth rate of traffic load, serving as the basis for subsequent regulation. For example, in monitoring a city's main road, analyzing the changes in average daily traffic flow over the past three years yields an r value of 0.12, indicating that the road's traffic flow is increasing at an annual growth rate of 12%.

[0106] S502: Setting an upper limit for land carrying capacity Where A is the area of ​​the region, and α is the land carrying capacity coefficient, the value of which is generally determined based on actual land resources and planning standards, for example, it can be set to 0.8-1.2, ultimately requiring a match between actual land use and infrastructure. This formula is used to quantify the maximum traffic demand capacity that a region can carry. For example, in the construction of a new urban area, according to the planning drawings, the area A = 50 square kilometers, and α = 1.0, we get K = 500,000 square meters, indicating that the area can accommodate a maximum traffic flow of 500,000 square meters.

[0107] S503: Assess the congestion risk in the current area Where D_t represents real-time traffic flow and K represents the carrying capacity limit, this formula combines traffic volume growth and current utilization rate to calculate the potential risk level. (β is the warning threshold) When subsequent measures are triggered. For example, in the operation of a certain road segment, D_t=450,000, K=500,000, r=0.12, after substituting, we get M=0.108. If β=0.1, then the road segment has not yet exceeded the critical point, but it still needs to be continuously monitored.

[0108] S504: The proportion of reserved resources to be allocated is determined by combining the ratio of M and K. Where δ is the control coefficient, typically ranging from 0.5 to 1, representing the adjustment range. This formula indicates that as the risk of congestion increases, more traffic resources need to be allocated. For example, in the case above, if M=0.108, β=0.1, δ=0.8, R=1-(0.108 / 0.1)×0.8=0.136, it means that approximately 13.6% of additional resources need to be invested for dynamically adjusting traffic allocation.

[0109] This technical solution enhances the intelligence level of land and space management by linking time-series data and resource allocation mechanisms, reduces the probability of sudden congestion events, and ensures the efficient and sustainable development of the transportation system.

[0110] Next, refer to Figure 6 The steps for generating a dynamic reservation scheme according to the present invention are described below.

[0111] S601: Establishing a Traffic Demand Growth Model Where T_b is the time period number, used to represent the time-series information of different periods. Parameters a and b are fitting coefficients, with a typically ranging from 0.5 to 1.5 and b from -1 to 1, to reflect the linear and nonlinear effects of time on traffic demand. This model aims to simulate the changing trend of traffic flow over time, ensuring that the prediction results match the actual situation.

[0112] S602: Based on the degree of land use change This formula assesses the activity level of spatial changes. U_t represents the land use type variable at time t, and its value can be a discrete number; for example, 0 represents residential land, and 1 represents commercial land. A larger C_p value indicates more frequent changes in spatial use. This formula can be used to evaluate the stability of regional dynamic changes, supporting more refined spatial planning decisions.

[0113] S603: Input the model output Y into the dynamic allocation function Here, γ is an adjustment parameter, which is recommended to be within the range of 0.5 to 2 to balance the impact of traffic demand and spatial activity. This formula integrates these two factors to obtain a comprehensive index, providing a basis for subsequent traffic resource allocation.

[0114] S604: Traffic demand is stratified based on Q-values, with higher-level demands given priority in response. For example, in a city expansion area, as new developments progress, T_b increases from 1 to 5. At this point, U_t changes significantly, causing C_p to rise, which in turn drives up the Q-value. The system will include roads and bus routes with higher Q-values ​​in the priority optimization scope, improving traffic efficiency and public service supply in that area.

[0115] This technical solution can effectively link traffic demand with changes in land use, improve the intelligence level of national spatial planning, enhance the ability to respond to emergencies, improve resource utilization efficiency, and help form a scientific and precise dynamic control mechanism.

[0116] Next, refer to Figure 7 This describes the steps of using the prediction results of a spatiotemporal evolution model to perform temporal regulation of transportation facility corridor space according to another embodiment of the present invention.

[0117] S701: Analysis of Road Utilization Rate Indicators Where v_i represents the number of vehicles within a certain time period, v_ideal is the ideal traffic volume, and N is the number of time segments. The purpose of this formula is to quantify the actual utilization efficiency of the current road segment. The range of v_i depends on the road segment capacity, v_ideal is calculated from traffic flow theory, and a lower L_u value indicates lower efficiency. For example, on a main road during peak hours, v_i reaches its limit, and L_u will increase significantly.

[0118] S702: Establishing Resource Allocation Equations Where L_w is the road segment width, L_a is the average traffic density, and t_w is the time weight, representing the importance of a time period. This formula measures resource utilization by integrating multiple factors; a higher R_d indicates a greater degree of resource strain. For example, if an increase in L_w and t_w during weekday morning and evening rush hours causes an increase in R_d on a certain expressway, it is necessary to consider whether traffic management strategies need to be optimized.

[0119] S703: Resource gaps are assessed based on R_d and prioritized. When R_d is greater than 1, it indicates that existing resources are insufficient to meet demand, and an expansion plan should be arranged; when R_d is less than or equal to 1, current conditions meet operational requirements, and no emergency intervention is needed. This dynamic adjustment method helps optimize resource allocation and improve system operating efficiency. For example, after a new residential area is built in a certain region, increased traffic pressure causes R_d to exceed the threshold, requiring adjustments to lane settings to ensure smooth travel.

[0120] This method combines multi-dimensional data analysis to achieve dynamic monitoring and precise decision-making of transportation facility corridor space, reduce traffic congestion risks, and improve the efficiency of land use. It has high practicality and operability.

[0121] Next, refer to Figure 8 This describes the process of building a spatiotemporal evolution model according to an embodiment of the present invention.

[0122] S801: Collect spatial information across multiple dimensions, including road network structure S_r, land use type U_l, and the proportion of different modes of transportation M_t. This step aims to acquire core spatial data related to national land spatial planning. S_r represents the road network layout and its connectivity, U_l refers to different types of land use attributes, such as residential, commercial, and industrial land, and M_t describes the proportion of different modes of transportation within the transportation system. For example, in one embodiment, road network, land use classification maps, and public transportation passenger flow data for a city over many years can be collected. By integrating this dimensional information, comprehensive spatial characteristics can be formed.

[0123] S802: Calculate the similarity matrix between each dimension. This step uses a cosine similarity algorithm to compare the degree of correlation between data from different dimensions, reflecting their interaction in space and time. Cosine measures the similarity of the angle between vectors, with values ​​ranging from -1 to 1; a value closer to 1 indicates higher similarity. For example, in a case study analyzing the relationship between land use and traffic distribution, it can be found that commercial land and bus stops have a high similarity, indicating a strong correlation between the two.

[0124] S803: Constructing a weight matrix by incorporating time-varying factors The exponential part of this formula is used to characterize the dynamic trend of spatial features changing over time. δ controls the rate of change and is generally set to a positive value. Its optimal value can be adjusted according to the time sensitivity of the study area. t represents the current time point, and t_0 represents the initial time point. For example, in urban renewal projects, land use patterns change slowly in the early stages and gradually accelerate in the later stages; this formula can capture this gradual process.

[0125] S804: Combining similarity with time weights forms a multi-dimensional fusion feature matrix A=M_sim×W_t, used for modeling and analysis. This matrix reflects both the correlation between data in each dimension and their significant changes over different time periods, providing dynamic support for land spatial planning. For example, during the development of a certain area, A can reflect changes in road network adjustments and industrial land layout in real time, assisting the government in optimizing management strategies.

[0126] In a specific implementation, after matrix multiplication to obtain A, it needs to be structured to adapt to the model input. Each row of matrix A corresponds to a spatial analysis unit with a unique ID (such as a 500m grid), and each column represents a feature that integrates specific spatial relationships and temporal weights. For example, the first column might be "timeliness similarity between road network structure and land use type," and the second column might be "timeliness similarity between land use type and traffic mode proportion." These feature columns together constitute the spatiotemporal feature profile of each spatial unit.

[0127] Secondly, matrix A is used as the feature dataset (X), and paired with historical traffic flow data for the corresponding period (as the target variable y) to form a training sample set. Then, a suitable machine learning or deep learning model is selected for training. For example:

[0128] Tree ensemble models (such as XGBoost and LightGBM) are used: The feature columns of matrix A are directly input into the model. The model can automatically learn the importance of different features and establish complex decision rules such as "when the time-similarity feature value of 'road network-land use' increases significantly in the near future (with high weight), the probability of future traffic flow growth increases".

[0129] Using spatiotemporal sequence models (such as LSTM and Transformer): If the feature vectors of each spatial unit across multiple consecutive time periods (i.e., the sequence of changes over time in that row of matrix A) are arranged in chronological order, a spatiotemporal sequence sample can be constructed. Recurrent neural networks or attention mechanism models can capture the dynamic patterns of these fused features evolving over time, thereby enabling more accurate temporal extrapolation predictions.

[0130] Graph Neural Networks (GNNs) are employed: spatial units are viewed as nodes in a graph, spatial adjacencies or functional connections between units are considered as edges, and the attributes of each node are its feature vectors in matrix A. GNNs aggregate neighborhood information through message passing, enabling them to model spatial dependencies, predict traffic distribution across the entire region, or identify areas that are about to become hotspots due to surrounding influences.

[0131] Furthermore, it can be used for unsupervised learning and pattern discovery. Without pre-setting any objectives, cluster analysis is directly performed on all row vectors of matrix A (e.g., using K-means or DBSCAN algorithms). Its physical significance lies in grouping regions with similar spatiotemporal collaborative development patterns into one category. For example, it might discover a category of regions characterized by "moderate but continuously increasing road network-land use similarity, and a recent rapid increase in the weight of diversified transportation modes." This type of region might correspond to "industry-city integration zones undergoing transformation," whose transportation demand growth pattern is unique. This pattern discovery provides a direct basis for formulating differentiated, categorized corridor reservation policies.

[0132] Finally, the model is used for interpretability analysis and dynamic iteration. The predictions of a model trained on the A matrix can be interpreted using methods such as feature importance analysis, partial dependency graphs, or SHAP values. Planners can clearly understand which specific spatiotemporal correlation factors (such as "the recent matching degree between road network and commercial land use") dominate the model's judgments, thus corroborating data-driven conclusions with professional planning knowledge and enhancing decision-making confidence. Furthermore, the A matrix is ​​a dynamic data product. With the arrival of new year's data, the system automatically updates the t-value, recalculates W_t and M_sim, generates a new A matrix, and uses this to fine-tune or retrain the model. This makes the entire modeling and analysis system a continuously learning and dynamically evolving system that accompanies urban development, ensuring the timeliness and forward-looking nature of planning recommendations.

[0133] In addition, such as Figure 9 As shown, the present invention also provides a land spatial planning management system 900 based on big data, comprising:

[0134] Data acquisition module 901 collects historical traffic flow data and land use evolution data, and performs preprocessing and standardization.

[0135] The model building module 902 constructs a spatiotemporal evolution model based on the historical traffic flow and land use evolution trends and identifies potential traffic growth hotspots.

[0136] The control module 903 performs temporal control on the transportation facility corridor space based on the prediction results of the spatiotemporal evolution model and generates a dynamic reservation scheme.

[0137] Feedback module 904 feeds back the dynamic reservation scheme to the land and space planning system to guide the timing of infrastructure construction.

[0138] The functions of each module of the big data-based land spatial planning management system 900 of this invention have been referenced above. Figures 1-8 The description has already been provided, and will not be repeated here.

[0139] This invention provides a big data-based land spatial planning and management method, comprising: First, acquiring historical traffic flow data and land use evolution data through a multi-source data acquisition system, and preprocessing and standardizing the data to ensure consistency and usability. Then, utilizing data analysis and mining techniques, combined with spatiotemporal modeling methods, constructing a spatiotemporal evolution model based on historical traffic flow trends and land use evolution characteristics, thereby identifying potential traffic growth hotspots. This model can effectively predict future changes in traffic demand and land function conversion trends in specific areas. Based on this, dynamically and temporally controlling the spatial layout of transportation facility corridors according to the prediction results, and formulating reasonable construction schedules and reserve plans for transportation infrastructure. This dynamic reserve plan, by feeding the prediction results back to the land spatial planning system, achieves scientific planning and efficient scheduling of infrastructure construction, avoiding resource waste and planning errors caused by blindly building ahead of schedule. Specifically, through in-depth analysis and trend forecasting of historical data, it is possible to accurately identify which regions will face significant traffic congestion in the near future. Based on this, the construction pace of transportation infrastructure can be optimized, ensuring that various infrastructure projects are implemented at the optimal time. This achieves optimal infrastructure allocation in land spatial planning and solves the technical problems of premature infrastructure construction and resource waste that may occur with traditional methods. This approach not only improves land resource utilization efficiency but also enhances the flexibility and foresight of planning, possessing significant practical application value.

[0140] The methods, programs, systems, apparatuses, etc., in embodiments of the present invention can be executed or implemented in one or more networked computers, or practiced in a distributed computing environment. In the embodiments of this specification, in these distributed computing environments, tasks can be performed by remote processing devices connected via a communication network.

[0141] Those skilled in the art will understand that the embodiments described in this specification can be provided as methods, systems, or computer program products. Therefore, those skilled in the art will realize that the functional modules / units or controllers and related method steps described in the above embodiments can be implemented in software, hardware, or a combination of both.

[0142] Unless explicitly stated otherwise, the actions or steps of the methods and procedures described in the embodiments of the present invention do not necessarily have to be performed in a specific order and can still achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0143] This document describes several embodiments of the present invention; however, for the sake of brevity, the descriptions of the embodiments are not exhaustive, and identical or similar features or parts between the embodiments may be omitted. In this document, "one embodiment," "some embodiments," "example," "specific example," or "some examples" refers to embodiments applicable to at least one, but not all, of the present invention. The above terms do not necessarily refer to the same embodiments or examples. Without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of the different embodiments or examples.

[0144] The exemplary systems and methods of the present invention have been specifically shown and described with reference to the above embodiments, which are merely examples of the best mode for implementing the systems and methods. Those skilled in the art will understand that various changes can be made to the embodiments of the systems and methods described herein without departing from the spirit and scope of the invention as defined in the appended claims when implementing the systems and / or methods.

Claims

1. A big data-based national space planning management method, characterized in that, include: Historical traffic flow data and land use evolution data are collected, and preprocessed and standardized. Based on the historical traffic flow and land use evolution trends, a spatiotemporal evolution model is constructed, and continuous spatial clusters with traffic demand growth rates consistently higher than the average level are identified as potential traffic growth hotspots. Based on the prediction results of the spatiotemporal evolution model, the space of transportation facility corridors is subject to temporal regulation and a dynamic reservation scheme is generated. The dynamic reservation scheme will be fed back into the land and space planning system to guide the timing of infrastructure construction. in The spatiotemporal evolution model constructed based on historical traffic flow and land use evolution trends includes: Obtain the traffic flow data matrix D for each region within different time periods; Obtain the land use type distribution matrix U for the corresponding time period; The data is extracted by a sliding time window, and a time series correlation coefficient matrix is calculated where N is the number of sampling points, and D^T is the transpose of the traffic flow matrix The threshold Δ is used to determine whether a change in land use structure has occurred: if max(C) > Δ, then proceed to the next stage; and The temporal regulation of transportation facility corridor space based on the prediction results of the spatiotemporal evolution model includes: Calculate the current regional traffic load rate ρ = actual traffic flow / expected maximum traffic volume; Construct the time decay factor λ=e^(-t / ), where t is the time step from the current time. The preset adjustment time constant; The reserved space weight W = (ρ-0.5)×(1-λ)+λ is calculated based on ρ and λ. When W is greater than the set threshold η, it is marked as a priority allocation area; otherwise, allocation is postponed. The step of feeding the dynamic reservation scheme back to the land and space planning system further includes: Generate a traffic corridor reservation level sequence ...,Pn}, each level corresponds to a different construction time window length; The weighted entropy model S=-Σ(p_i×log(p_i)) is used for scheme optimization and selection, where p_i represents the probability of a certain level being selected; The final dynamic scheme to be retained is determined based on the size of S. The scheme with the smallest S is the optimal one. If the actual construction progress deviates from the scheme by more than ε, the readjustment process is initiated.

2. The land spatial planning management method based on big data according to claim 1, characterized in that, The temporal regulation of transportation facility corridor space further includes: Extract the historical traffic flow growth rate r = (D_t+1-D_t) / D_t, where D_t is the traffic flow data at time t; Set the upper limit of land carrying capacity K = A × α, where A is the area of ​​the region and α is the land carrying capacity coefficient; Determine the current congestion risk in the area as M = r × (D_t / K); when M > β, initiate the dynamic reservation procedure, where β is the warning threshold. The proportion of reserved resources is determined by the ratio of M and K: R = 1 - (M / β) × δ, where δ is the control coefficient.

3. The land spatial planning management method based on big data according to claim 1, characterized in that, The dynamic reservation scheme for generation further includes: Establish a traffic demand growth model Y = a × T_b + b × ln(T_b), where T_b is the time period number; The degree of spatial change activity is determined by the land use change rate C_p = Σ(U_t+1-U_t)^2, where U_t represents the land use type variable in period t. The model output Y is input into the dynamic allocation function Q=(Y+C_p)×γ, where γ is the adjustment parameter; Traffic demand is categorized into different levels based on the Q value, with higher-level demands given priority in response.

4. The land spatial planning management method based on big data according to claim 3, characterized in that, The temporal regulation of transportation facility corridor space further includes: The road utilization rate index L_u = ∑(v_i^2 / v_ideal) / N is analyzed, where v_i is the number of vehicles in a certain period of time and v_ideal is the ideal traffic volume. Establish the resource allocation equation R_d=(L_u×L_w) / (L_a×t_w), where L_w is the width, L_a is the average density, and t_w is the time weight; Resource gaps are assessed based on R_d and prioritized. If R_d > 1, it means that expansion is needed; otherwise, there is no need to intervene immediately.

5. The land spatial planning management method based on big data according to claim 1, characterized in that, The construction of the spatiotemporal evolution model further includes: Collect spatial information from multiple dimensions, including road network structure S_r, land use type U_l, and the proportion of different modes of transportation M_t; Calculate the similarity matrix M_sim=[cosine(S_r,U_l),cosine(U_l,M_t)] between each dimension; A weight matrix W_t=exp(-δ×(t-t_0)) is constructed by combining the time-varying factor, where δ controls the rate of change and t_0 is the starting time; By combining similarity with time weights, a multidimensional fusion feature matrix A=M_sim×W_t is formed for modeling and analysis.

6. A land spatial planning management system based on big data, characterized in that, include: The data acquisition module collects historical traffic flow data and land use evolution data, and performs preprocessing and standardization. The model building module constructs a spatiotemporal evolution model based on the historical traffic flow and land use evolution trends, and identifies continuous spatial clusters with traffic demand growth rates consistently higher than the average level as potential traffic growth hotspots. The control module performs temporal control on the transportation facility corridor space based on the prediction results of the spatiotemporal evolution model and generates a dynamic reservation scheme. The feedback module feeds the dynamic reservation scheme back to the land and space planning system to guide the timing of infrastructure construction. in The spatiotemporal evolution model constructed based on historical traffic flow and land use evolution trends includes: Obtain the traffic flow data matrix D for each region within different time periods; Obtain the land use type distribution matrix U for the corresponding time period; Feature extraction is performed on the data using a sliding time window, and the time-series correlation coefficient matrix is ​​calculated. , where N is the number of sampling points, and D^T is the transpose of the traffic flow matrix; The threshold Δ is used to determine whether a change in land use structure has occurred: if max(C) > Δ, then proceed to the next stage; and The temporal regulation of transportation facility corridor space based on the prediction results of the spatiotemporal evolution model includes: Calculate the current regional traffic load rate ρ = actual traffic flow / expected maximum traffic volume; Construct the time decay factor λ=e^(-t / ), where t is the time step from the current time. The preset adjustment time constant; The reserved space weight W = (ρ-0.5)×(1-λ)+λ is calculated based on ρ and λ. When W is greater than the set threshold η, it is marked as a priority allocation area; otherwise, allocation is postponed. The step of feeding the dynamic reservation scheme back to the land and space planning system further includes: Generate a traffic corridor reservation level sequence ...,Pn}, each level corresponds to a different construction time window length; The weighted entropy model S=-Σ(p_i×log(p_i)) is used for scheme optimization and selection, where p_i represents the probability of a certain level being selected; The final dynamic scheme to be retained is determined based on the size of S. The scheme with the smallest S is the optimal one. If the actual construction progress deviates from the scheme by more than ε, the readjustment process is initiated.

Images