WP 1 – Project Management
This task is transversal to the whole project and has as main objective to ensure a sound and efficient management of the SIM4SECURITY research.
WP 2 – Study, analysis and diagnosis of the current national situation
This task aims to perform a full diagnosis of the national state of the art concerning population and public security, namely: population dynamics, security forces (resources and location), characterization of the distribution and typology of the security forces according to population characteristics.
WP 3 – Demographic Forecast and Scenario development
In this WP, the team will develop a demographic Forecast and Scenario development based on population risk groups. This database and projections will be executed from 2011 to 2030 by parish disaggregation, and incuding scenarios by sex, age and micro simulation of risk groups (according to previously identified criteria of age, sex, national origin, place of residence) scenarios. Finally a Distribution Model of the Security Forces strength and equipment location will be developed.
WP 4 – Development and Implementation of a Geographic Information System and design of a dynamic geoprocessing Model
This task is transversal to all the former tasks, since its main goal is to developed and implement a GIS Model and application, which will support all the geographic and alphanumeric information produced during this research project.
But why should we use predictive and dynamic models? First of all, a model has the capability to support a decision or design process in which a user wishes to find a solution to a spatial problem, perhaps a solution that optimizes some objective, by giving the possibility to experiment on a world replica. Lastly, a model allows users to examine dynamic outcomes by viewing the modeled system as it evolves and responds to inputs. Scenarios evaluated with dynamic models are thus a very effective way of motivating and supporting debates over policies and decisions.
The SIM4SECURITY Model will have to comply with a set of rules and procedures in order to represent and predict a specific outcome. This model, integrated and supported by the WebGIS application will allow users to easily simulate, emulate and handle impacts generated by possible changes, where a simple change of a parameter, variable or factor will provide distinct dynamic
scenarios, visible on a map, aiming to assist planning issues and location-allocation problems related to the public security sector, such as: 1) suitability of the police offer distribution according to population characteristics; 2) study of risk groups spatial and temporal dynamics and impact assessment in police distribution; 3) deploying police forces; 4) estimation of public security needs for a given area; 5) number and type of professionals. All of these examples have spatial expression and they will differ according to the developed scenarios. This application will try to solve some location-allocation problems, issues that involve two types of decisions: where to locate and how to allocate demand for a service. This stands as one of the great advantages in using predictive models with spatial interaction.
The GIS application, which will be available in a restricted and secure web application, will be the key interface between users and the model, in which, besides using basic spatial visualization, navigation and spatial tools, will also gave them the freedom to manipulate and change parameters/factors, allowing them to visualize its changes, effects and impacts produced on a dynamic map.
This task will have as a main output the development and implementation of a Web-based GIS application and its Model.
WP 5 – Implementation of Advanced Spatial Analysis Methods (spatially dynamic clusters and modeling land cover change predictive model)
Several studies about land use change have been developed with the Land Transformation Model (LTM). This is an LUCC model based on GIS, Artificial Neural Networks (ANN) routines, remote sensing and geospatial analysis tools. LTM provides the dynamic modelling of social, political and environmental factors, such as the distance to public transportation and road networks,
proximity to natural resources such as rivers and lakes, agricultural and forest densities, identification of exclusionary zones and population growth.
LTM is an LUCC model based on ANN routines that “learn” about complex spatial relationships of factors that correlate with urban development. The ANN are employed in studies about urban growth because they learn about the relationships that exist between urban growth factors and the site attributes. LTM studies demonstrated that dynamic modelling and the scenario prediction are essential to planning and territorial management. Due to LTM land use modelling and forecasting capabilities, we decided to use this model in this study.
In this study we will model urban growth over the study area using the LTM. The urban areas will be obtained from CORINE Land cover (CLC) map of years 2000 and 2006. The LUCC model will use distance to roads, slope and distance to city centers as drivers of urban growth. Another important driver that will be used is the demographic growth studied in task 3. The model will be validated using percent correct metric and the kappa statistic. Projections of urban growth will then be made for years 2010, 2020 and 2030.
In both task 3 and in the first part of this task geospatial information will be outputted. In the first case, a Demographic Forecast and some Scenario development will be presented while in the last, forecast maps of land use for the next 10, 20 and 30 years will be produced. In the second part of this task it is our goal to combine these two data sets allowing a more detailed analysis of
the population spatialtemporal.
Since the demographic forecast will be aggregated by administrative regions (parish) it assumes that any outputted statistics have a homogeneous distribution. To get a higher detailed this is combined with urban versus nonurban maps allowing heterogeneous administrative regions regarding the forecasted variables. The goal here is to create spatially contiguous or nearcontiguous regions using several criteria such as the resident population, socioeconomic attributes or total area. To achieve this goal, we will apply two different approaches. In the first approach an artificial neural network called SOM will be used to cluster the input data into spatially consistent regions of similar socioeconomic attributes. Several restrictions can be added to this optimization analysis, such as the total area or the number of inhabitants per region.
The second approach will take advantage of evolutionary computation methods, more specifically Genetic Algorithms (GA) to present several possible regional grouping solutions. GA have been successfully used in optimization tasks and are considered one of the most powerful optimization strategies available. Work developed by the research team (Roberto Henriques) includes the application of GAs to zone design problems. Nevertheless, GA remains unexplored in this field and further research is necessary to develop an appropriate model.
The output of this task will be a clustering model capable of creating regions where similar security approaches should be undertaken easing the definition and implementation of policy measures.
WP 6 – Modeling the distribution of Security Forces (number of officers and facilities location) according to the developed scenarios
This task has the goal to create a distribution and location allocation model for the distribution of Security Forces (both number of officers and facilities location), according to the developed scenarios.
This task will try to solve some locationallocation problems, issues that typically involve two types of decisions: where to locate and how to allocate demand for a service. (25), taking into account several factors, the number of facilities available, their cost, and maximum impedance from a facility, for instance. This stands as one of the great advantages in using predictive models
with spatial interaction, such as the proposed application and model, in which our research team has relevant experience. The SIM4SECURITY GIS Model and GIS application aims to create several locationallocation scenarios and population forecasting for 2030.
For this task, the team project will use geographic and alphanumeric information created and collected during tasks 2, 3 4 and 5, which will allow to contribute to conduct studies that may gave some hints and clues on how to make a better and effective security resources management and allocation, considering the diversity and specificity of each area, crime levels and public perceptions about security.
It aims to contribute to the creation of a symmetric and optimized distribution of the public security resources throughout the territory, considering the population distribution, optimizing the racio police / citizen.
For these studies, the use of the SIM4SECURITY Model (created during the task 4 activites), will be crucial, since it allows users to dynamically simulate, emulate and handle impacts generated by scenarios and population distribution, aiming to assist planning issues and beforementioned location-allocation problems, regarding the internal security sector.
Different partners will be involved in this task. IPRI will bring their expertise on spatial planning policies, social aspects and demographics. IPRI and SSI will contribute with their experience and knowhow regarding internal security issues. Finally, NOVA IMS, will contribute with the SIM4SECURITY Model developed during task 4.
WP 7 – Project outreach and visibility
The main objective of this task is to disseminate the results of the project, both along its development and its final results.
This will be done through three components: the edition of two dedicated books (ebook and Hardcover), the organization of a workshop by the beginning of the project and of a international Seminar by the end of it. One of the main outputs, nevertheless will consist on the conception and development of a project website, whose advantages for public security will be formally presented in the previously refereed Seminar.