### Assessment for CASA0002 Urban Simulation

report | 代写Network | 代做network | mining – 这是利用report进行训练的代写, 对report的流程进行训练解析, 涵盖了report/Network/network/mining等程序代做方面

#### 15

### Urban Simulation

**Part 1: Londons underground resilience**

#### 1. Topological network

In most cases, a graph model of big metropolitan street layout is provided. The vertices and edges of this shape indicate streets and intersections, respectively. Street connectivity, characteristic path duration, and clustering factor are all used in structural analysis. A k- clustering coefficient is used to quantify the degree of clustering in a street network by taking into account the k-neighbours of each street in the network. Scale-free properties aren’t found in three cities where validations were carried out by specialists, according to experts. In this paper, I will evaluate the resilience of the London underground using the following procedures;

#### I. Centrality measures

To comprehend networks or graphs, centrality measurements are essential. Nodes’ relative importance in a network can be calculated using graph theory. Although these work slightly differently, they can all help identify areas of a network that require improvement. You need to know how each metric defines ‘importance’ for your graph visualization applications to choose the ideal one for your needs. Looking at a graph or network, you’ll want to look for "centrality" indicators that identify the virtual nodes. Because the definition of "importance" can vary from person to person, specific nodes may be considered more significant than others. Distinct flavors of centrality have different definitions of a node’s importance. Hence experts can measure each flavor’s centrality differently. Due to its wide range of applications, centrality is frequently the first metric taught to newcomers to network analysis. Experts can identify an intranet network’s important infrastructure nodes or disease spreaders by calculating the influence coefficient. The halfway measures are; Degree centrality, closeness centrality, and Betweenness centrality. A city’s ties and systems appear to be complicated, and the relationships within and between them are often unclear. Nearly the last century, the percentage of the world population residing in urban areas has climbed from 3 percent to over 50 percent. Rapid technological advancement has increased the size and complexity of cities, exacerbated by urbanization (Grawe et al., 2012). There are many advantages to the current trend of urbanization, but there are also several drawbacks. City benefits and problems scale super- linearly with the city’s size, and an increasing worldwide urban population is indeed exacerbating the hidden rivalry between urban improvement and degradation. The subject of how urban characteristics interact and relate to one another has become more critical than ever. Business and daily life in the city are interconnected because of the city’s extensive network of transportation systems. Because of the rise in personal mobility, the city’s existing public transportation infrastructure is straining to keep up with demand while still providing the high-quality services that citizens have grown to expect. Understanding how the current system’s features relate to (impact and be affected by) urban human geography is one technique to understand the demand for public transportation better.

#### A. Complex model of transport

Analysis of transportation costs and benefits, efficiency, and redundancy can be done using several approaches. One popular option is to use network science techniques to model complicated systems with unambiguous node and relational link definitions (Rossomakha, 2020). Transport systems have undergone substantial network and graph theory examination, both within and between cities, for highways, railroads, maritime commerce, and air travel. As far back as 2002, Boston’s subway system was characterized, while London’s metro system was described more recently. The London Underground’s core-periphery topology, for example, has been identified as an intriguing network property. Uncertainty remains regarding how transportation network characteristics relate to urban human geography characteristicsidentifying the prevalent correlations (without directly inferring causality) between urban dwelling areas. Research conducted over the previous 20 years has already revealed the importance of transit and urban elements, such as property prices and the experience of traveling through surveys.

#### 1. Degree centrality measure

D(s) = deg for a station’s degree, the number of links to other stations (s). This study emphasizes the number of ties rather than focusing on a neighbor or link (Sotoodeh & Falahrad, 2019). The term "hub" refers to stations with considerable dispersion.

#### 2. Closeness centrality measure

The number of hops it takes to get from one node to another in a graph is known as the "farness" of a node. The reciprocal of farness is the centrality of closeness C(s). From all other stations in the network, it’s possible to think of how quickly one can go to a specific station. Green Park has the shortest travel time to other stations, whereas Lewisham has the longest.

#### 3. Betweenness centrality measure

In order to minimize the number of station stops, commuter routes often use the shortest multi-hop route between stations. The shortest route between two points, m, and n, is the one that requires the fewest stops along the way (hops). The number of shortest routes that cross through a station Bs is meant by its "betweenness." In other words, B(s) =m=n=s

m,n(s), which is the shortest path from the station s to station n.

*It’s depicted in Figure 2. Underground and Overground Rail Networks shown to a certain extent (connectivity). the centrality of betweenness (shortest paths). (c) the significance of proximity (distance). centrality of eigenvectors (influence). PageRank (normalised influence). f) Group (local network).*

*In Fig. 2, each neighbourhood is centred on the award and averages the centrality scores of all stations in its region.* Impact measures This review summarizes the most critical findings from the research on the effects of node and link removal in genuine social networks. We only consider applications that use the two most widely used measures of network robustness, i.e., the most significant connected component (LCC) and network efficiency, for our evaluation (Eff). Both a binary and a weighted network method are on the table. We demonstrate that exa mining how social networks respond when links or nodes are removed is a highly effective tool for dealing with various real-world issues. For example, we show that the effects of social distancing in various states to restrict the spread of COVID-19 may be evaluated using social network analysis. Assess the weight of a person’s connections in a social network while thoroughly analyzing. Finally, we identify future research avenues in social network analysis. The term "network assault analysis" refers to research that has examined how real networks respond to the loss of links or nodes (LNR), simulating an attack just on the network. A wide range of sectors, including biology, ecology, transportation and infrastructure, information technology and neuroscience, and economics and social networks, all benefited from these investigations. Network robustness, a metric that measures how well a system can continue to function after an LNR, was one goal of these investigations. The other was to find out which LNRs were causing the most system damage and, as a result, which links/nodes were playing critical roles in network performance in each case. Focusing on LNR in genuine social networks, we describe relations between individuals, groups, organizations, societies, etc. We will present the most critical findings from the research and explain how they might be applied in practice.

#### i. Cluster centrality measure

With cluster centrality (local), we can see that the stations with the highest rankings create a distinct local network with other stations, which is a whole different interpretation of

centrality (i.e., triangles).

#### ii. Eigenvector centrality measure

Eigenvector centrality measures a station’s impact on the entire network. Alternatively, it could be read as the proximity to other important stations in the network rather than any particular set of stations being close to each other (YU & WANG, 2009). Accumulated through the jacobian matrix A = [A-S-0] (a binary relation among nodes), where the betweenness centrality of the node at the center is:

Alternately, one might write the adjacency matrix in the adjacency matrix equation AE=E, where E is the adjacency matrix, and E is an adjacency matrix equation that only applies to positive values of the adjacency matrix (E). Using PageRank, we may scale the eigenvector centrality by the number of stations each station has as neighbors. In terms of both eigenvector and PageRank, Bank comes out on top. The two measures are not specific to London alone; simulators can use them to measure and evaluate the resilience of any other network. For every sector of society, the effects of an internet outage are becoming more and worse. As a result, future networks must be made more resilient and durable. Network resilience is defined as the ability to continue providing intended service even in the face of assaults, disasters, and other problems (Veitch & Johnson, 1997). nodes can improve the Future Internet’s resilience and survivability by using a blend of topology generation, analytical, modeling, and experimental emulation techniques, which this article covers.

#### II. Node removal

There may be a correlation between the removal of nodes and information propagation in social networks regarding individual abandonment. It can help identify the essential network nodes with various interpretations. It’s possible to remove a node from a network using node deletion, either by chance or directly. Networks can be tested for their resilience and attack tolerance by removing nodes from the network. In many empirical networks, it is crucial to understand how a network changes due to node elimination.

#### A. Non-sequential

When the doors open, the red and amber signal lamps system is activated regardless of whether or not the amber lamps have been engaged. It is known as Non-Sequential Operations.

[ndbd default] DataMemory = 2 00M IndexMemory = 2 00M NoOfReplicas = 4 DataDir = /usr/local/mysqr/val/mysql-cluster [ndbd] Id = 2 HostName = 198.51.100. [ndbd] Id = 4 HostName = 198.51.100. [mgm] HostName = 198.51.100.

Id = 40 [api] Id= 80 HostName = 198.51.100. [api] Id= 42 HostName = 198.51.100.

#### B. Sequential removal

It’s possible to organize records sequentially in a sequential file. The records are arranged in a predetermined order. It is impossible to read or write records in a sequential file other than in logical order. You can’t cut, copy, or delete a record once it’s been added to a sequential file. — NDB Cluster — service to a customer– ndb_mgm> SHOW Admin Servers are connected.: 200. 50. 200. 05 : 1986 Clusters Configurations

[ndbd(NDB)] 2 node(s) id=1 @198. 51. 100. 1 ( 5. 7. 37 – ndb- 7. 5. 26 , Nodegroup: 0 , *) id=2 @198. 51. 100. 2 ( 5. 7. 37 – ndb- 7. 5. 26 , Nodegroup: 0 )

[ndb_mgmd(MGM)] 1 node(s) id=10 @ 200. 50. 200. 05 : 1986 ( 5. 7. 37 – ndb- 7. 5. 26 )

[mysqld(API)] 2 node(s) id=20 @ 200. 50. 200. 05 : 1986 ( 5. 7. 37 – ndb- 7. 5. 26 ) id=21 @ 200. 50. 200. 05 : 1986 ( 5. 7. 37 – ndb- 7. 5. 26 ) The best strategy is degree centrality measure. When calculating degree centrality it’s one of the simplest. The number of edges a node has determines its degree of centrality. The more important a node is, the greater its degree. Since so many nodes having high degrees both have high central by other measures, this can be an essential method. A vertex’s degree centrality is just a measurement of the total number of connections. It can be viewed as a crude form of popularity measurement, but it doesn’t differentiate between amount and quality in terms of its results. Linking a recent trainee to an organization’s CEO doesn’t make any difference to the degree of centrality. Edges connecting to a vertex are counted as a vertex’s "degree." In-degree and out-degree are two metrics of degree in directed networks, where linkages have an origin and a goal rather than mutual connections. There are a certain number of connections in each vertex that point inward, known as the in-degree. Several connections from a vertex to other vertices are referred to as out-degrees.

#### 2. Flows: weighted network

weighted network type of the weighted network, where the ties between nodes have been assigned a specific weight. A network is a system in which all parts are somehow linked together (Mironov, 2007). Nodes and links (also known as ties, edges, arcs, or arcs) represent a system’s components. Individuals, organizations, airports, and even countries are all examples of nodes, whereas linkages might be friendship, communication, cooperation, alliance, flow, or trade. The measure best fits the passengers, so I cannot remove or adjust it.

*Fig. 3. Weighted networks in London* Degree centrality is the most critical measure in London, and this is because it is easy to calculate in different stations. The stations are given below;

*Fig. 4. The London metro system’s 15 most vital stations are vulnerable to attack. Passengers must travel 1.6 kilometers to reach the next-door station. A station’s inoperability is calculated using the nonnumbers parentheses. (Color version available online.)*

ndb_mgm> ALL report MEMORY

Node 1: Data usage is 13%( 180 32K pages of a total of 3200) Node 1: Index usage is 11 %( 80 8K pages of a total 12832) Node 2: Data usage is 13%( 180 32K pages of a total of 3200) Node 2: Index usage is 11%( 75 8K pages of a total 12832)

Node 3: Data usage is 3%( 79 32K pages of a total of 3200) Node 3: Index usage is 1%( 53 8K pages of a total 12832) Node 4: Data usage is 3%( 81 32K pages of a total of 3200) Node 4: Index usage is 1%( 52 8K pages of a total 12832)

**Part 2: Spatial Interaction models**

#### III. Models and calibration

It is common to utilize spatial interaction models to estimate trip creation and distribution as the first two phases in a four-step transportation and land-use model. According to many spatial interaction models, flows are influenced by a combination of the properties of origin and destination and a measure of the distance between the two points. A spatial interaction model can be summarized as follows:

```
Where;
```

*Tij*: Interaction between I (the source) and j (the destination) (destination). Quantities such as passengers, tons of freight, and traffic volume can all be used to measure it. Encounters by the hour, day, month, or year are examples of this.*Vi*: Social and economic aspects of the place of origin, such as population, job availability, factory production, or any proxy for the number of economic activities such as gross domestic product (GDP), are commonly employed to convey these features.*Wj*: Features of the location where destination j will be located This feature employs the same socioeconomic variables as the previous one to emphasize the comparability of the areas.*Sij*: The characteristics of the distance between the origin and the destination. Transport friction, distance friction, or impedance are other names for this concept. Distance, transportation expenses, and journey time are standard metrics for describing these properties. Complementarity is expressed in the most excellent feasible way when the qualities of V and W are paired. As an example, one might use the working-age population (V) and total employment (W) to estimate commuting flows (work-related movements).**The gravity model**Using the gravity model is the most popular approach to spatial interaction in computer science. Because of the similarity in formulation, it is known as Newton’s law of gravitation. When it comes to analyzing the boundaries between different markets, gravity-like representations have been used in various circumstances (Sohn & Yoon, 2000). A relationship between two items’ mass and distance is inversely related to their attraction. I can construct a simple gravity model by starting with a broad description of spatial interactions that incorporates this fundamental assumption:

```
Where;
```

*Pi*and*Pj*are important for both the origin and destination locations.

- did; Distance between the starting point and the final destination.
- The rate of the occurrence is proportional to the proportionality constant, k. Considering the same network of spatial interactions over a year yields a higher k value than considering the same system over a week would yield a higher k value. When two sites are of equal importance, their distance is proportional to the amount of interaction between them. It’s possible to extend the gravity model to incorporate several additional calibration parameters:

It is difficult to calibrate spatial interaction models, such as the gravity model, which poses a considerable obstacle to their use. The simulator must calibrate each parameter in the model (constant and exponent) to guarantee that the estimation results are consistent with the observed flows, can be repeated, and can be generated by altering the parameters. An incorrect model is useless if it does not accurately forecast or explain what will happen in the future. It is hard to tell if the calibration procedure is effective without comparing the outcomes of the estimates with the actual results. For the model to be as accurate as possible, calibration must be done consistently.

```
Fig. 5. Spatial Interactions are affected by beta, alpha, and lambda.
```

*Figure 6. After Brexit, Canary Wharf had a 50% drop in employment.* Those with less education and poorer health were more likely to leave the workforce early because of reduced employment protections. Workers with less education and lower health may benefit from employment protection regulations, which may lessen their disadvantage in the workplace. A supply chain’s international transportation component is reduced when supply sources move closer to end consumers, reducing distance-driven expenses. In addition, sourcing from near-shore sources, such as North America, Latin America, and the Caribbean, has further advantages. Additionally, by importing goods from Latin America and the Caribbean through East Coast ports, shippers can avoid congestion on the U.S. West Coast and avoid the more expensive cross-country moves from West Coast ports to population centers east. Reduced port congestion and delays imply that enterprises can respond more rapidly to changes in client demand because of the shorter distance. As a result, this transition is helpful from both a financial and a customer service standpoint.

References Grawe, D., Thompson, H., Salmond, J., Cai, X., & Schlnzen, K. (2012). Modeling the impact of urbanization on regional climate in the Greater London Area. *International Journal Of Climatology* , *33* (10), 2388-2401. https://doi.org/10.1002/joc. Mironov, A. (2007). Weighted graphs with fixed vertex degrees and network flows. *Journal Of Computer And Systems Sciences International* , *46* (3), 413-417. https://doi.org/10.1134/s Rossomakha, O. (2020). Conceptual model of the system of maintenance and repair of complex technical systems. *Transport Development* , *6* (1), 56-70. https://doi.org/10.33082/td.2020.1-6. Sohn, C., & Yoon, J. (2000). (Gravity Model)) (A Gravity Model Analysis of Korea’s Trade Patterns and the Effects of a Regional Economic Bloc). *SSRN Electronic Journal*. https://doi.org/10.2139/ssrn. Sotoodeh, H., & Falahrad, M. (2019). Relative Degree Structural Hole Centrality, CRDSH: A New Centrality Measure in Complex Networks. *Journal Of Systems Science And Complexity* , *32* (5), 1306-1323. https://doi.org/10.1007/s11424-018-7331- Veitch, P., & Johnson, D. (1997). ATM network resilience. *IEEE Network* , *11* (5), 26-33. https://doi.org/10.1109/65. YU, W., & WANG, J. (2009). Suspicious money laundering detection system based on eigenvector centrality measure of transaction network. *Journal Of Computer Applications* , *29* (9), 2581-2585. https://doi.org/10.3724/sp.j.1087.2009.