Using spatial point process models, clustering and space partitioning to reconfigure fire stations layout

Abstract. Geospatial approaches are widely used to organize access and to manage the extinguishing of forest fires globally. Term “transport accessibility” is used in a variety of geographical and economic researches. Assessment of transport accessibility is directly related to the feasibility study to locate the fire stations in a particular region. Location analysis of objects relative to other objects, while taking into account various quantitative and qualitative parameters, is a classical problem solved by geoinformation systems.Present research work is aimed to be used to improve the situation with forest fires in Russia where one of the main asset of operational regional firefighting in the forests is a fire-and-chemical (fire) station. Traditionally station placement is under the responsibility of Russian region to which stations are administratively subordinate. The location of fire station is determined taking into account the species structure of forests, natural fire danger, road infrastructure and some other factors. Irkutsk region, one of the territories with the constant perennial fire danger in the forests, was chosen as a test area.Using this area as a typical example of regional extent, an analysis of fire stations placement then planning the ground movement of fire brigades from a station to the forest fire locations has been carried out. Previously obtained results to create the shortest routes within three-hour accessibility for the fire hazardous seasons 2002–2017 are used. Russian regulatory documents of the forest industry are applied. Thus, topic of GIS analysis serves as a continuation of the study (http://cepl.rssi.ru/wp-content/uploads/2018/11/Podolskaya-E.S.-et.-al..pdf, in Russian) and various aspects of transportation problem are considered on the example of Irkutsk region.We used the following input data: point layer of fire stations, road network (including roads of different classes and forest glades), and archive of forest fires detected using the spectroradiometer MODIS from the Aqua and Terra spacecraft. Additional data were collected from the open regional Internet sources. GIS analysis used ArcGIS ArcMap Desktop extensions such as Network Analyst, Spatial Analyst, and ArcGIS tools like ET GeoWizards (https://www.ian-ko.com/).We have developed and have used a forest fire transport model of ground access by trucks for the Irkutsk region with the spatial arrangement of fire stations, two protection zones and road network. Speed of the forest fire trucks is classified into 5 groups, taking into account the official permissions and road class. Also every segment of road has its attributive data of speed “adjusted” by the elevation value of the ETopo2, an open-access model (https://www.ngdc.noaa.gov/mgg/global/etopo2.html). Taking into consideration these relief data allows to decrease the vehicle’s speed in the mountainous conditions.Based on the regional specifics and available data for a fire hazardous season, the following set of evaluation parameters was proposed, namely:Road network: roads existence, length, density, and configuration;Spatial distribution of detected forest fires;Territory of fire stations servicing.All the listed parameters are interconnected to each other and, in combination, jointly impact the stations ground transport accessibility assessment at the regional level. We have used GIS-analysis methods such as buffering, allocation, and density, as well as geographic and directional distribution. Time frame of analysis is the full fire hazardous season. Undertaken analysis for the forest fires detected within the ground protection zone (archive of 2002–2017) has shown that the fire stations’ distribution was appropriate. The analysis was based on the assumption that stations had the same weight, geographical and transport location of the stations was reviewed in conjunction to the thematical forestry recommendations. To go further in the GIS analysis some characteristics (work force and technical resources) of the weighted stations could be added.Additional factors of influence can be the location of protected areas with their specific access regime, seasonality of road use, forest fire zoning, forestry boundaries, economic criteria, placement of fire stations in the populated area, etc. It is advisable to conduct a fire stations placement analysis as a preparation event before and after the end of the fire-hazardous season to summarize the effectiveness of actions to extinguish forest fires in the region. Practical results of the study can be used as well to prepare the regional forestry development programs and plans.

Cellular networks have a great transformation during the last decade. The traditional cellular network system is often overlaid with well-deployed BSs of proper power. However, more and more complex elements such as picocells, femtocells and relays have been introduced to cellular network system. BSs in these elements are unplanned, user-installed and randomly deployed. All of these things compose heterogeneous cellular networks (HCNs). In order to evaluate the performance of HCNs, people usually model the distribution of BSs in HCNs as spatial stochastic point process model. One of the most popular spatial models for these HCNs is Poisson Point Process (PPP) model. However the PPP model is not exact for some cases where BSs in PPP model may be too close to each other. In this paper, we propose a more reasonable and practical model called Matern-like point process (MLPP) model which imposes a certain minimal distance between any two BSs to overcome this problem. Simulations show that the coverage probability of the proposed model noticeably outperforms the PPP model. This improvement depends on the value of minimal distance. We also evaluate the system efficiency defined by the ratio of coverage probability and total power. The results show that the MLPP model has a significant improvement compared to PPP model.

GIScience 2016 Short Paper Proceedings Location Optimization of Fire Stations: Trade-off between Accessibility and Service Coverage J. Yao 1 , X. Zhang 2 Urban Big Data Centre, University of Glasgow, 7 Lilybank Gardens, Glasgow, G12 8RZ, UK Email: Jing.Yao@glasgow.ac.uk Department of Geographical Information Science, Hohai University, 1 Xikang Road, Nanjing, 210098, China Email: Xiaoxiang@hhu.edu.cn Abstract Fire and rescue service is one of the fundamental public services provided by government in order to protect people, properties and environment from fires and other disasters, and thus promote a safe living environment. Efficient deployment of fire stations is necessary and essential if timely response to the emergencies is to be achieved. Spatial optimization approaches have been long employed in public facility location studies. In particular, coverage-based models, such as the location set covering problem (LSCP) and the maximum coverage location problem (MCLP), have been widely adopted to achieve complete or maximum coverage of service demand. This paper extends the LSCP by accounting for both partial coverage and access to the demand areas. The proposed model is applied to the optimization of fire station locations in Nanjing, China. The results can be used to assist future fire station location planning and rescue resource deployment. 1. Introduction Fire caused by humans or nature can pose hazard to people, properties and environment, and lead to psychological damage, physical injuries (even death) and significant economic losses. Fire prevention and protection is necessary and essential for a safe living environment. The associated fire and rescue service therefore needs to be properly deployed to ensure efficient fire safety management. A fundamental concern in this regard is the spatial configuration of fire stations as it is critical to timely response to emergency calls. Given the inherent spatial nature, fire station location problems have been well studied using geographical information system (GIS)-coupled location modelling (Chevalier et al. 2012; Aktas et al. 2013; Murray 2013). In particular, LSCP (Toregas et al. 1971), MCLP (Church and ReVelle 1974) and their extensions have long been employed to evaluate the locational efficiency of existing fire stations as well as seek sites for new fire stations (Chevalier et al. 2012; Murray 2013). Common goals of locating fire stations include maximizing the access to provided services, covering as much demand as possible and minimizing total costs of service provision, usually subject to available resources. In practice, two or more objectives are often considered to capture different aspects in relation to fire service delivery. The aim of this paper is to seek best locations of fire stations with spatial optimization approaches, particularly considering accessibility and service coverage. The proposed model is applied in an empirical study in Nanjing, China, to assist future fire station location planning and rescue resource deployment.

The geographical distribution of the population of the city of Santiago, Chile, has changed significantly in recent years. In spite of this fact, the location of the fire stations has remained unchanged. We propose a model for the optimal location of the fire stations and a fleet assignment for the Santiago Fire Department (SFD), aimed at maximising the number of events attended to with a predefined standard response. The results of the model are compared with respect to the current location of fire stations and fleet assignment in the SFD. There are different types of resources (stations and vehicles), and different types of events in which the same types of vehicles are used. We analyse various possible current and future scenarios, using a forecast based on historical data. Our results show that by optimally reallocating the resources a 10–30% increase can be achieved in the number of emergency calls that are attended to with an adequate response in time and number of vehicles, without the need for additional fire stations or vehicles. Thus our contribution is empirical and relies on the real world application which is being considered by Chilean government.

As the global aviation business expands rapidly, issues of aviation safety become correspondingly important. In turn, aviation safety should be more emphasized. The crashes of China Airlines planes at Nagoya (Japan) international airport in 1994, and near Taipei's international airport in 1998, caused airport authorities around the world to pay closer attention to rescue and fire‐protection plans at their airports. Our research reveals that the location and number of fire stations at an international airport is an important factor in its fire protection capability. However, if the sites of the fire stations are not appropriately planned and located, fire engines and crews cannot arrive at the accident area in a timely manner. Similarly, if the number of fire stations at an airport is not sufficient, fires caused by aircraft accidents may take longer to be extinguished, resulting in more injuries and fatalities. Therefore, a location model based on a fuzzy multi‐objective approach is proposed in this paper. This model can help in determining the optimal number and sites of fire stations at an international airport, and can also assist the relevant authorities in drawing up optimal locations for fire stations. Finally, because of the combinatorial complexity of our model, a genetic algorithm (GA) is employed and compared with the enumeration method. The study results show that our revised GA is comparatively effective in resolution and that our model can be applied to the optimal location of other emergency facilities.

Two classes of ergodic stationary multitype spatial point processes are constructed. These processes have the property that interactions between the types exist, but cannot be detected using standard second-order methods of analysis. Simulations indicate that the interactions can, however, be detected by using ‘empty space' techniques.

Introduction. The article presents the results of a study of the capabilities of the Vietnamese fire service to respond to various destructive events using the KOSMAS computer simulation system (Emergency Services Computer Modeling System), and also presents and evaluates possible options for the development of fire service forces and means and their deployment in Vietnam. Goals and objectives. Study of the capabilities of the Vietnamese fire service to respond to various destructive events using simulation technologies. Methods. Simulation (computer) modelling. Results and discussion. By means of a simulation model, an assessment of the coverage by the response time of fire departments of all buildings and structures located on the territory of Vietnam is given and the number and locations of the new fire stations are determined. Conclusions. As a result of the simulation experiments, in addition to the existing fire stations, 11 new potential locations for fire stations were identified, the commissioning of which will reduce on 2-3 min average response time of the first fire units in the country. Taking into account the new fire stations, 87.4 % of all buildings and structures available on the territory of the country will be located in the 20-minute response time zone of the fire units (that is 6.4 % more compared to the existing option) and almost 94.4 % – in the 30-minute response time. Keywords: response time, fire stations, operational activities, simulation model, simulation system.

One challenge in managing fire hazards in an urban setting is how to optimize the fire service route, increase the response time, and increase service coverage. Recently, this challenge is becoming imminent due to road traffic congestion and insufficient road widths that are common in populated cities in the Southeast Asia regions. One of the urban fire hotspots in populated Jakarta City is Pekojan Urban Village, Tambora Subdistrict. This subdistrict is served by Angke Fire Station located in Pekojan’s southwestern parts. Then this research aims to evaluate and compare optimized routes for fire vehicle dispatched from Angke Fire Station to serve 12 neighborhood units (in Bahasa is RW) in Pekojan. The method used the route optimization and network analysis tools in Geographic Information System (GIS) and its related geospatial data including neighborhood units, road networks, traffic congestion, and fire station locations. Geospatial network analysis of data by GIS has an advantage as a method to design and analyze the routing strategy and determine the most optimized route for fire vehicles. Based on the results and with the fire vehicle speed of 40 km/h, the average optimized route distances to travel from the fire station to RWs were 1.092 km (95%CI: 0.888-1.3 km) with an average response time of 1.638 minutes (95%CI: 0.869-2.41 min.). According to the GIS, model, response time of 1 minute only covers 22.77% of Pekojan areas. By increasing response time to 2 minutes, then fire vehicle can cover 98.9% of Pekojan area (AIC= 0.06). Despite the fact that the fire vehicle routes and response times can be optimized, those routes are challenged by the road traffic congestion. This congestion limits the speeds of fire vehicles to less than 20 km/h, as observed in 11.59% of the optimized routes. The service coverages of fire vehicles was also limited due to the narrow street.

Failure to account for time‐varying detection ranges when inferring space use of marine species from passive acoustic telemetry data can bias estimates and result in erroneous biological conclusions. This potential source of bias is widely acknowledged but often ignored in practice due to a lack of available statistical methods. Here, we describe and apply a spatial point process model for estimating individual centres of activity (COAs) from acoustic telemetry data that can be modified to account for both receiver‐ and time‐specific detection probabilities. We use simulation testing to evaluate the suitability of the proposed models for estimating COAs and compare their performance to that of the popular mean‐weighted COA method for a variety of scenarios. We illustrate how the approach can be applied to correct for variable detection ranges by integrating data from moored test tags and demonstrate how accounting for time‐varying detection probabilities can impact space use estimates by fitting the model to data collected from a black sea bass (Centropristis striata) on a receiver array off the east coast of the United States. The proposed model reduced bias in COA estimates, particularly when tagged individuals occurred along the periphery of the receiver array. The test tag‐integrated model largely corrected the bias associated with receiver‐ and time‐specific detection probabilities. When applied to the black sea bass detection data, the model revealed fine‐scale movements not apparent when detection ranges were assumed constant. Spatial management practices for coastal marine species are often based on trends in space use inferred from passive acoustic telemetry data, which can be misinterpreted when factors influencing detection ranges are not accounted for. Our approach provides a general framework for estimating individual COAs that can be modified on a study‐specific basis to ensure resulting patterns of space use reflect a species’ movements and behaviour, rather than variation in receiver detection ranges.

For a spatial point process model in which the intensity depends on spatial covariates, we develop graphical diagnostics for validating the covariate effect term in the model, and for assessing whether another covariate should be added to the model. The diagnostics are point-process counterparts of the well-known partial residual plots (component-plus-residual plots) and added variable plots for generalized linear models. The new diagnostics can be derived as limits of these classical techniques under increasingly fine discretization, which leads to efficient numerical approximations. The diagnostics can also be recognized as integrals of the point process residuals, enabling us to prove asymptotic results. The diagnostics perform correctly in a simulation experiment. We demonstrate their utility in an application to geological exploration, in which a point pattern of gold deposits is modeled as a point process with intensity depending on the distance to the nearest geological fault. Online supplementary materials include technical proofs, computer code, and results of a simulation study.

This paper summarizes two studies conducted to improve the emergency service delivery in St. Albert, a small city near Edmonton, Alberta. The first study dealt with selecting the location of a new fire station. The goals of the study included an assessment of the performance of the current system, an identification of the area(s) with poor coverage, a selection of a site among a set of given candidate locations, and an assessment of the improvement in the system performance upon the addition of the new fire station. A geographical information system was used for storing and displaying the spatial data, computing service areas for given travel times, and for communicating the results of the study. The second study considered an evaluation of the resources available for emergency service. A probabilistic model was used to evaluate labor costs for different platoon sizes and a simulation model was used to evaluate the adequacy of the current staff and fleet sizes. Results were presented to EMS officers, city staff, as well as the City Council.

We propose and investigate dynamic spatial-temporal point process models for independent and interacting events. The models for independent events are dynamic spatial-temporal Poisson point process (DSTPPP) model that account for temporal and spatial clustering. The models proposed for events with interaction are Markov (Gibbs) space-time point process models. We model the intensity function of a DSTPPP via conditioning arguments that allow for additional interpretations and inclusion of well-known point process models as special cases. Depending on the nature of the questions to be answered, the in- tensity function of a DSTPPP can be modeled in several ways. First, we develop models via conditioning on the time component of the events and then consider conditioning on the location component of the events. Modeling, simulation and computation are accomplished in a fully hierarchical Bayesian framework. Finally, we focus on dynamic marked Markov space-time point processes, where the events are allowed to interact with each other across time and space. Once again the hierarchical Bayesian framework is invaluable in this case, since it allows us to introduce dynamic process models via conditioning. The methodolo- gies are illustrated using simulated data, and several applications, including modeling and inference for earthquake events in California, and tornado events in Missouri.

The positive association between alcohol outlet density and assault rates is well established, but little is known about how this association differs across victim subpopulations. We use spatial point process models on police data from Flint, Michigan, to test how the link between alcohol outlet density and assault rates changes as a function of three victim characteristics: age, gender, and race. We found that, although both on-premises and package outlet densities consistently emerge as risk factors for victimization, their relative effects are markedly larger in Whites than in African Americans. No analogous age- or gender-based differences were found. These results suggest the racial effects arise more from relative differences in the atmosphere in and around alcohol outlets than differences in drinking behavior alone.

Measuring the spatial accessibility to fire stations using enhanced floating catchment method

ABSTRACTIn this paper, we derive an exact formula for the covariance of two innovations computed from a spatial Gibbs point process and suggest a fast method for estimating this covariance. We show how this methodology can be used to estimate the asymptotic covariance matrix of the maximum pseudo‐likelihood estimator of the parameters of a spatial Gibbs point process model. This allows us to construct asymptotic confidence intervals for the parameters. We illustrate the efficiency of our procedure in a simulation study for several classical parametric models. The procedure is implemented in the statistical software R, and it is included in spatstat, which is an R package for analyzing spatial point patterns.