Identification Method for the Interaction Mode between Residents' Commuting Behavior and Land Use Imbalance

Abstract

The present invention discloses a method for identifying the interaction mode between residents' commuting behavior and land use imbalance, including: collecting and processing the research area dataset; dividing the research area into several traffic analysis zones (TAZs); calculating the directed commuting attraction between any two TAZ pairs based on the gravity model, using the TAZs as the nodes of the bipartite network, and using the value of the directed commuting attraction as the weighted edge between the nodes of the bipartite network to construct a weighted bipartite network; mapping the actual commuting OD flow data to the weighted bipartite network to obtain the number of job positions and the number of residents of each node in the weighted bipartite network, and further determining the commuting flow capacity between any two TAZs; performing optimal flow allocation with the optimization goal of maximizing the global edge weight of the network; comparing the actual and optimal commuting flows in the network to identify two abnormal commuting modes of significant overload and significant underload; and outputting the identification result of the abnormal commuting mode.

Description

Technical Field

[0001] The present invention belongs to the technical field of geographic information science, and particularly relates to a method for identifying the interaction mode between residents' commuting behavior and land use imbalance. Background Art

[0002] In recent years, the rapid expansion of cities has greatly changed the urban spatial structure. The abnormal interaction between the urban spatial structure and residents' activities has brought problems such as job-housing imbalance, traffic congestion, and low land use efficiency, seriously threatening the sustainable development of cities. Commuting is the main travel activity of big city residents. Due to the unreasonable spatial layout of urban infrastructure, especially for vulnerable groups, residents have to face a heavy commuting burden. Some areas have concentrated excessive commuting populations, while in some areas, the commuting flow has not reached the expected level. In order to achieve fair and sustainable urban planning, it is of great significance to identify the imbalance interaction mode between residents' commuting activities and urban land use functions.

[0003] The interaction modes between commuting and land use imbalance can be classified into two categories: overload and underload. The commuting overload imbalance interaction mode means that the total commuting flow in this area exceeds its carrying capacity, which often occurs in old urban areas or old industrial parks that have not undergone urban renewal. The commuting underload imbalance interaction mode means that this area has a perfect infrastructure system, but there are not enough people working in this area, which usually occurs in new urban areas that have not gathered enough popularity. For example, in order to rapidly expand the built-up area, some cities have developed a large number of infrastructure projects in the suburbs, which is very likely to lead to the idle of public resources.

[0004] The commuting flow network is an effective method for identifying abnormal commuting patterns. However, current research based on the commuting flow network often focuses on the network's own structure and statistical characteristics. On the one hand, it quantifies network structure indicators such as the connectivity and stability of the network itself. On the other hand, it analyzes statistical quantities such as the job-housing balance index and employment demand in the network, and then distinguishes different commuting patterns. Such methods pay too much attention to the aggregation and sparsity of the commuting flow and are difficult to effectively reflect the interaction process between commuting behavior and urban land use. In addition, when optimizing the distribution of the commuting flow, taking the minimum commuting time or distance as the goal ignores the efficiency of the urban land use function, which may lead to the inconsistency between the optimized commuting flow and the actual service range of urban infrastructure.

[0005] Therefore, how to determine the optimal commuting interaction flow between any two units in the commuting flow network and how to identify the two interaction modes of overload and underload between residents' commuting behavior and land use imbalance still need to be broken through. These related issues are also the key to deeply understanding the interaction mode between commuting behavior and urban land use space and improving urban commuting and land use efficiency. Summary of the Invention

[0006] The object of the present invention is to provide a method for identifying the interaction pattern between residents' commuting behavior and land use imbalance in view of the deficiencies of the prior art. The present invention solves the optimal commuting flow between urban units from the perspective of optimizing the interaction network traffic, effectively reflecting the interaction process between residents' commuting behavior and land use. Moreover, the present invention can identify two abnormal commuting patterns of overloading and underloading, improving the data coverage, uniformity and the validity of the conclusion.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions:

[0008] A method for identifying the interaction pattern between residents' commuting behavior and land use imbalance includes the following steps:

[0009] Step 1: Collect and process the dataset of the study area. Among them, the collected data includes the big data of residents' commuting OD in the study area;

[0010] Step 2: Divide the study area into several traffic analysis zones (TAZs);

[0011] Step 3: Calculate the directed commuting attraction between any two TAZ pairs based on the gravity model, and use the TAZs obtained in Step 2 as the nodes of the bipartite network, and use the value of the directed commuting attraction as the weighted edge between the nodes of the bipartite network to construct a weighted bipartite network;

[0012] Step 4: Map the actual commuting OD flow data to the weighted bipartite network to obtain the number of job positions and the number of residents of each node in the weighted bipartite network, and then determine the commuting flow capacity between any two TAZs;

[0013] Step 5: Under the constraint conditions, take maximizing the global edge weight of the network as the optimization objective for optimal flow allocation;

[0014] Step 6: Compare the actual and optimal commuting flows in the network to identify two abnormal commuting patterns of significant overloading and significant underloading;

[0015] Step 7: Output the recognition result of the abnormal commuting pattern and display its spatial distribution.

[0016] Furthermore, the big data of residents' commuting OD in Step 1 includes the coordinates of the user's place of residence, the coordinates of the workplace, the commuting distance, the commuting time, the Baidu user heat data in the study area, and the road network data in the study area. Among them, at least one end of the user's place of residence and place of employment is located in the study area.

[0017] Furthermore, the method for processing the data in Step 1 is as follows:

[0018] 1) Unify the coordinate system of the spatial vector data to form a vector dataset with consistent geographic coordinate system and projected coordinate system;

[0019] 2) Clean the abnormal commuting flow data within the spatial scope of the study area according to the commuting time and commuting distance fields.

[0020] Furthermore, the method for calculating the directed commuting attraction in step 3 is as follows:

[0021] For any two TAZs i and TAZ j , the directed commuting attraction A i of TAZ j to TAZ ij is calculated by the gravity model as:

[0022]

[0023] In the formula, P i is the Baidu user heat density of TAZ i during the day; Dis ij is the road network distance between TAZ i and TAZ j ; β is the scaling factor used to adjust the influence degree of accessibility on the directed commuting attraction; F i is the functional level related to commuting employment of TAZ i , which is calculated from the density of different categories of POIs:

[0024]

[0025] In the formula, ω k represents the weight of the kth type of POI category, determined according to the commuting survey data; Dp k represents the density of the kth type of POI category; NP represents the total number of POI categories.

[0026] Furthermore, the commuting flow capacity between any two TAZ nodes v i and node u j in step 4 is determined as:

[0027]

[0028] In the formula, RE i is the resident population of node v i , EM j is the number of employment positions of node u j . The employment position information and resident population information of the nodes are obtained by summarizing the resident commuting OD big data to the TAZ, and do not include the residents who live and work within the same TAZ.

[0029] Furthermore, the specific implementation method of step 5 is as follows:

[0030] If S and T are regarded as virtual source and sink nodes with infinite capacity, and it is assumed that all commuting flows flow out from S and flow into T, then under the constraint conditions, the optimal commuting interaction flux between TAZs is expressed as a maximum matching problem:

[0031]

[0032] The constraint conditions are:

[0033]

[0034] where f ij is the commuting flow from node v i to node u j ; f si is the commuting flow from the virtual node S to node v i , that is, the flow from the place of residence; f jt is the commuting flow from node u j to the virtual node T, that is, the flow to the place of employment;

[0035] The three constraint conditions respectively represent the capacity constraint between nodes, no provincial flow in the network, and the flow is proportional to the weight of the edge.

[0036] Furthermore, in step 5, the Hungarian algorithm based on capacity update is used for optimal flow allocation.

[0037] Furthermore, the allocation method of the Hungarian algorithm is as follows:

[0038] First, use the Hungarian algorithm for the original network to obtain the initial residual network G r ; by subtracting the allocated resident population and employment positions, update the new capacity between any two nodes in this network. At the same time, when the capacity between nodes is empty, set the weight between nodes to 0, and record the updated situation of the commuting flow between nodes in the flow allocation network G f ;

[0039] Based on the updated edge weights, use the Hungarian algorithm again to find a new maximum matching, and repeat this process until the capacity between all nodes in the residual network G r is empty; finally, output G f as the optimal commuting flow allocation scheme.

[0040] Furthermore, in step 6, the abnormal commuting flow is calculated by the following formula:

[0041] AN(taz i →taz j ) = actual (taz i →taz j ) - optimal(taz i →taz j )

[0042] In the formula, F actual (taz i →taz j ) and F optimal (taz i →taz j ) respectively correspond to the actual commuting flow and the optimal commuting flow from taz i to taz j ; AN(taz i →taz j ) is the abnormality degree of the commuting flow. A positive value means that the actual commuting flow is too high compared with the optimal commuting flow, and "commuting congestion" will occur. A negative value indicates that the actual commuting flow is lower than the optimal commuting flow, and there is a problem of insufficient commuting in the area. Positive and negative values respectively represent two abnormal commuting modes of overload and underload.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0044] From the perspective of optimizing the interactive network traffic, the present invention solves the optimal commuting flow between urban units, overcoming the problem that the prior art cannot determine the optimal commuting interaction flow between any two units in the commuting flow network. In addition, the present invention quantifies the attractiveness of the region from three aspects: traffic conditions, population vitality level, and POI facilities, measures the impact of the urban spatial structure on residents' commuting, and effectively reflects the interaction process between residents' commuting behavior and land use; with the maximum global land use attractiveness as the optimization goal, it fully considers the efficiency of the urban land use function; driven by commuting OD big data, it avoids the complicated on-site investigation process in the early stage, and improves the data coverage, uniformity and the effectiveness of the conclusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Overall flowchart of the method for identifying the imbalance interaction mode between residents' commuting behavior and land use in the embodiment of the present invention;

[0046] Figure 2 Flowchart of the Hungarian algorithm based on capacity update in the embodiment of the present invention;

[0047] Figure 3 Spatial distribution map of the optimal commuting interaction flow in the embodiment of the present invention;

[0048] Figure 4 Spatial distribution map of the overload commuting interaction mode in the embodiment of the present invention;

[0049] Figure 5 Spatial distribution map of the underload commuting interaction mode in the embodiment of the present invention. [[ID=5

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0052] The present invention will be further described below in conjunction with specific embodiments, but it is not limited to the present invention.

[0053] To solve the problems existing in the prior art, an identification method for the interaction mode between residents' commuting behavior and land use imbalance is provided in the embodiments of the present invention. This method uses traffic analysis zones (TAZs) as the analysis unit, constructs a framework based on a weighted bipartite network to detect abnormal commuting interactions. In the process of constructing the bipartite network, first, the attraction between TAZ pairs is measured based on the gravity model as the weighted edges between TAZs in the bipartite network; then, an improved Hungarian algorithm is used to solve the global maximization of the edge weights under the flow capacity constraint to obtain the optimal commuting interaction flow between TAZs; finally, by comparing the actual and optimal commuting interaction flows, two types of abnormal commuting modes, overloading and underloading, are identified, and on this basis, measures and suggestions for urban spatial renewal and commuting flow allocation are proposed. See Figure 1 , the embodiments of the present invention specifically include the following steps:

[0054] Step 1: Collect and process the research area dataset. Among them, the collected data includes the large data of residents' commuting OD in the research area;

[0055] In this embodiment, the main urban area of Wuhan is used as the research area, and Baidu commuting OD data is collected. The dataset contains the coordinates of users' residential places, the coordinates of workplaces, commuting distances, and commuting times; the Baidu user heat data in the research area with a data resolution of 200m; the POI facility point data in the research area, including companies, government agencies and institutions, catering and entertainment, etc.; the road network data in the research area, including main and secondary roads, branch roads, ring roads, etc.

[0056] After obtaining the above data, the data is processed. The processing method is as follows:

[0057] (1) Perform unified projection conversion processing on the spatial vector data to form a vector dataset with a consistent spatial coordinate system;

[0058] (2) Clean the Baidu commuting OD data, delete the records outside the coordinate range of the main urban area of Wuhan, and set the commuting distance threshold to be between 0.5 km and 30 km, and the commuting time threshold to be between 1 minute and 4 hours.

[0059] Step 2: Divide the study area into several TAZs, that is, the areas enclosed by four main roads. After the preliminary division, adjust it in combination with the urban planning unit.

[0060] Step 3: Calculate the directed commuting attraction between any two TAZ pairs based on the gravity model, and use the TAZs obtained in Step 2 as the nodes of the bipartite network, and use the value of the directed commuting attraction as the weighted edge between the nodes of the bipartite network to construct a weighted bipartite network.

[0061] In this embodiment, the method for calculating the directed commuting attraction between any two TAZ pairs based on the gravity model is as follows:

[0062] The main factors attracting residents' commuting in the area include accessibility, regional population vitality, and facility functions. The accessibility of employment places is one of the decisive factors affecting residents' commuting behavior, and commuters tend to work near their place of residence; regional population vitality characterizes the ability of a region to carry human activities and reflects the trend of urban population flow; facility functions reflect the completeness of the supporting infrastructure related to commuting employment. Specifically, given two TAZs i and TAZ j , then the directed commuting attraction A i of TAZ j to TAZ ij can be calculated by the gravity model as:

[0063]

[0064] In the formula, P i is the Baidu user heat density of TAZ i during the day (7:00 - 21:00); Dis ij is the road network distance between TAZ i and TAZ j , which can be obtained based on the path planning API of Amap; β is a proportionality factor used to adjust the influence degree of accessibility on the directed commuting attraction; F i is the function level related to commuting employment of TAZ i , which can be calculated by the density of different categories of POIs:

[0065]

[0066] In the formula, ω k represents the weight of the kth type of POI category, which is determined according to the commuting survey data; Dpk Denote the density of the k-th type of POI; NP represents the total number of POI categories.

[0067] Step 4: Map the actual commuting OD flow data to the weighted bipartite network to obtain the number of job positions and the number of residents of each node in the weighted bipartite network, and then determine the commuting flow capacity between any two TAZ nodes v i and node u j where the commuting flow capacity between any two TAZ nodes v i and node u j is calculated by the formula:

[0068]

[0069] In the formula, RE i is the resident population of node v i EM j is the number of job positions of node u j The job position information and resident population information of the nodes can be obtained by summarizing the resident commuting OD data to the TAZ, and do not include the residents who live and work within the same TAZ. The number of job positions is also corrected by the scale factor.

[0070] Step 5: Under the constraint conditions, with the optimization goal of maximizing the global edge weight of the network, use the Hungarian algorithm based on capacity update for optimal flow allocation. The algorithm flow chart is as Figure 2 shown;

[0071] In this embodiment, S and T are regarded as virtual source nodes and sink nodes with infinite capacity, and it is assumed that all commuting flows flow out from S and flow into T. Then, under the constraint conditions, the optimal commuting interaction flux between TAZs can be expressed as a maximum matching problem:

[0072]

[0073] The constraint conditions are:

[0074]

[0075] In the formula, f ij is the commuting flow from node v i to node u j ; f si is the commuting flow from the virtual node S to node v i , that is, the flow from the place of residence; f jt is the commuting flow from node u j to the virtual node T, that is, the flow to the place of employment. The three constraint conditions respectively represent the capacity constraint between nodes, no provincial domain flow in the network, and the flow is proportional to the weight of the edge.

[0076] The Hungarian algorithm can provide a feasible flow allocation F that conforms to the maximum weight matching of the bipartite network f , so the improved Hungarian algorithm based on capacity update is used to calculate the optimal commuting interaction flux between TAZs. The calculation method is as follows:

[0077] First, use the Hungarian algorithm on the original network to obtain the initial residual network G r ; by subtracting the already allocated residential population and employment positions, update the new capacity between any two nodes in the network. At the same time, when the capacity between nodes is empty, set the weight between nodes to 0. The updated situation of the commuting flow between nodes is recorded in the flow allocation network G f ; based on the updated edge weights, use the Hungarian algorithm again to find a new maximum matching, and repeat this process until the capacity between all nodes in the residual network G r is empty; finally, output G f as the optimal commuting flow allocation scheme.

[0078] Step 6: Compare the actual and optimal commuting flows in the network to identify two abnormal commuting patterns of significant overload and significant underload. Among them, the abnormal commuting flow can be calculated by the following formula:

[0079] AN(taz i →taz j ) = actual (taz i →taz j ) - optimal (taz i →taz j );

[0080] In the formula, F actual (taz i →taz j ) and F optimal (taz i →taz j ) are the actual commuting flow and the optimal commuting flow from taz i to taz j respectively; AN(taz i →taz j ) is the abnormal degree of the commuting flow. A positive value means that the actual commuting flow is too high compared with the optimal commuting flow, and "commuting congestion" may occur. A negative value indicates that the actual commuting flow is lower than the optimal commuting flow, and there may be a problem of insufficient commuting in the area. Positive and negative values represent two abnormal commuting patterns of overload and underload respectively.

[0081] Step 7: Output the recognition results of abnormal commuting patterns and display their spatial distribution.

[0082] In this embodiment, based on the directed commuting attraction intensity, Figure 3 the spatial distribution of the optimal commuting flow allocation is shown. It can be found that the area with the highest commuting flow is not concentrated in the most central area of the city. On the contrary, the areas with the most frequent commuting interactions are located in the sub-centers of the city, such as Zhongnan Road, Luojia Road, etc. By comparing the actual commuting flow and the optimal commuting flow, abnormal commuting patterns are identified, and significant abnormal commuting areas are analyzed. As Figure 4 shown, three obvious overloaded commuting areas are identified, including Wuhan Iron and Steel Group (O1), Luoyu Road (O2), and South Lake (O3). The spatial distribution of the areas with overloaded commuting flow is similar to the areas where the actual commuting flow gathers, which can be explained from two aspects. On the one hand, since the distribution of commuting flow is proportional to the attraction level of multiple possibilities of destinations in each TAZ, to a certain extent, the commuting flow is dispersed. Therefore, when the actual commuting flow is large, the probability of overloaded commuting is high; on the other hand, the proper infrastructure construction in the city center provides strong commuting capacity, while the sub-city centers have limited employment capacity due to the high commuting demand in the surrounding areas, so commuting overload is likely to occur. Specifically, Wuhan Iron and Steel Group attracts a large number of long-distance commuters, but the infrastructure such as public transportation in this area is relatively weak, so it is more likely to have overloaded commuting. Similarly, the high commuting flow does not match the attraction in the areas near Luoyu Road and South Lake. In addition, some large residential communities where commuting needs are difficult to be met are found, and the surrounding urban building space needs to be renovated, such as Donghu Residential Area (H1) and Zhuanyang Avenue Residential Area (H2).

[0083] Commuting overload means that the commuting-related infrastructure in this area is underdeveloped, while commuting shortage indicates that this TAZ has a high attraction, but the commuting flow is lower than its carrying capacity. There are obvious differences between the two abnormal commuting patterns. As Figure 5 shown, the spatial distribution of the commuting underload pattern is more dispersed and has a smaller coverage area, which indicates that the commuting underload phenomenon is concentrated near high-attraction areas. In addition, the spatial distribution of commuting underload is highly correlated with the attraction level, but not completely consistent. Generally, most commuting underload interactions occur in the city centers (U1-U7) with insufficient employment attraction. The two abnormal commuting interaction patterns of commuting overload and commuting underload are often complementary in space, which is more obvious near Wuhan Iron and Steel Group (O1, U1) and Luojia Road (O2, U5). The spatial adjacency of the two abnormal commuting interaction patterns can help urban planning departments improve commuting efficiency, guide the reconfiguration of urban infrastructure, and the diversion of residents' commuting flows.

[0084] It should be understood that the parts not elaborated in detail in this specification belong to the prior art.

[0085] The above are only the preferred embodiments of the present invention, and thus do not limit the implementation manners and protection scope of the present invention. For those skilled in the art, it should be realized that all the equivalent substitutions and obvious changes made by using the content of the specification of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying the interaction mode between residents' commuting behavior and land use imbalance, characterized in that, It includes the following steps: Step 1: Collect and process the dataset of the study area. Among them, the collected data includes the big data of residents' commuting OD in the study area; Step 2: Divide the study area into several traffic analysis zones (TAZs); Step 3: Calculate the directed commuting attraction between any two TAZ pairs based on the gravity model, and use the TAZs obtained in Step 2 as the nodes of the bipartite network, and use the value of the directed commuting attraction as the weighted edge between the nodes of the bipartite network to construct a weighted bipartite network; Step 4: Map the actual commuting OD flow data to the weighted bipartite network to obtain the number of job positions and the number of residents of each node in the weighted bipartite network, and then determine the commuting flow capacity between any two TAZs; Step 5: Under the constraint conditions, perform optimal flow allocation with the optimization goal of maximizing the global edge weight of the network; Step 6: Compare the actual and optimal commuting flows in the network to identify two abnormal commuting patterns of significant overload and significant underload; Step 7: Output the recognition results of abnormal commuting patterns and display their spatial distributions; Among them, the method for calculating the directed commuting attraction in Step 3 is: For any two and , the directed commuting attraction to is calculated by the gravity model as follows: Wherein, is the Baidu user heat density during the day; is and the road network distance between; is the scale factor used to adjust the influence degree of accessible directed commuting on attraction; is the functional level related to commuting employment, which is calculated by the density of different categories of POIs: In the formula, represents the weight of the th type of POI category, which is determined according to the commuting survey data; represents the density of the th type of POI category; represents the total number of POI categories; Any two TAZ nodes in Step 4 and node The commuting flow capacity between them is determined as: In the formula, is the resident population of node , is the number of job positions of node . The job position information and resident population information of the nodes are obtained by aggregating the large data of residents' commuting OD to the TAZ, and do not include the residents who live and work within the same TAZ. The specific implementation method of Step 5 is as follows: If S and T are regarded as virtual source nodes and sink nodes with infinite capacities, and it is assumed that all commuting flows flow out from S and flow into T, then under the constraint conditions, the optimal commuting interaction flux between TAZs is expressed as a maximum matching problem: ； The constraint conditions are: ； where is the commuting flow from node to node ; is the commuting flow from the virtual node S to node , i.e., the flow from the place of residence; is the commuting flow from node to the virtual node T, i.e., the flow to the place of employment; The three constraint conditions respectively represent the capacity constraint between nodes, no provincial-level flow in the network, and the flow is proportional to the weight of the edge.

2. The identification method of the interaction mode between residents' commuting behavior and land use imbalance according to claim 1, wherein, The big data of residents' commuting OD in Step 1 includes the coordinates of users' residential places, the coordinates of workplaces, commuting distances, commuting times, the Baidu user heat data in the study area, and the road network data in the study area. Among them, at least one end of the user's residential place and employment place is located in the study area.

3. The recognition method of the interaction mode between residents' commuting behavior and land use imbalance according to claim 1, characterized in that, The method for processing the data in Step 1 is: 1) Unify the coordinate systems of the spatial vector data to form a vector dataset with consistent geographic coordinate system and projection coordinate system; 2) Clean the abnormal commuting flow data within the spatial range of the study area according to the commuting time and commuting distance fields.

4. The recognition method of the interaction pattern between residents' commuting behavior and land use imbalance according to claim 1, wherein In Step 5, the Hungarian algorithm based on capacity update is used for optimal flow allocation.

5. The identification method of the interaction pattern between residents' commuting behavior and land use imbalance according to claim 4, characterized in that The allocation method of the Hungarian algorithm is: First, use the Hungarian algorithm on the original network to obtain the initial residual network ; update the new capacity between any two nodes in the network by subtracting the already allocated resident population and employment positions. At the same time, when the capacity between nodes is empty, set the weight between nodes to 0, and record the updated commuting flow between nodes in the flow allocation network ; Based on the updated edge weights, use the Hungarian algorithm again to find a new maximum matching, and repeat this process until the capacities between all nodes in the residual network are empty; finally, output as the optimal commuting flow allocation scheme.

6. The recognition method of the interaction mode between residents' commuting behavior and land use imbalance according to claim 1, characterized in that The abnormal commuting flow in Step 6 is calculated by the following formula: In the formula, and correspond to the actual commuting flow and the optimal commuting flow from to respectively; is the degree of abnormality of the commuting flow. A positive value means that the actual commuting flow is too high compared with the optimal commuting flow, and "commuting congestion" will occur. A negative value indicates that the actual commuting flow is lower than the optimal commuting flow, and there is a problem of insufficient commuting in the area. Positive and negative values represent two abnormal commuting modes of overload and underload respectively.

Images