Allocation Of Police Resources Research Articles

Balancing act: Assessing police efficiency amidst staffing shortages through inverse data envelopment analysis

Abstract Police agencies are amid a significant staffing shortage. Scholars and police managers warn about the effects of understaffing on public safety and security. Studies have examined the impacts of events like the Floyd protests had on police resignations. This study uses a linear programming method, Inverse data envelopment analysis (InvDEA), to estimate the efficient allocation of police resources using a simulated agency with 150 sectors. The first decision-making scenario simulated the challenge many agencies currently face with staffing shortages. Reallocating officers and civilians from inefficient to efficient units improved performance by 4.7% without additional officers. Results of the second and third decision-making scenarios indicate a need to reallocate resources by 23% and 19%, respectively. These findings contribute to the scarce literature on police staff allocation. In addition, the methodology employed in this study is helpful for police managers to make accurate operational decisions when combined with changes to the efficiency frontier that reflect the environment in which they operate.

경찰의 초동대응 체계 개선을 위한 드론 활용 방안 연구

This study explores the necessity and potential of utilizing drone technology in police operations. To address the increasing crime rates and complex safety issues in modern society, the efficient allocation of police resources and swift response are becoming imperative. This research analyzes how drone technology can meet these demands and how AI can enhance the operation of drones from both theoretical and practical perspectives. The use of drones in first response scenarios can reduce arrival times at scenes, enable monitoring over extensive areas, and facilitate accurate situation awareness, thus supporting the police in taking quicker and more effective measures. However, the integration of such technologies necessitates consideration of the legal and ethical issues that may arise, as well as preventive measures against potential technological errors and privacy invasions. The study underscores the need for a legal basis, procedural legitimacy, and safe operational conditions when actively utilizing drones and AI technology in police activities, and presents specific recommendations for this purpose.

Rural policing in China: Criminal investigation and policing resources for police officers

This study examines the state of policing resources in rural areas of central China and investigates how police officers conduct criminal investigations. Semi-structured interviews were conducted with 20 police officers who had experience in criminal investigations in rural areas. The study found that police officers have developed a well-established model of criminal investigation that is effective in responding to minor crimes such as theft, drink driving, and fighting. However, there may be compliance issues with the criminal investigation skills of police officers. Additionally, rural areas lack policing resources compared to urban areas and do not have access to necessary police information resources. The results of this study suggest that rural policing in areas where resources are scarce should focus on improving the compliance of criminal investigation skills among police officers and increasing the allocation of police resources to rural areas. Policymakers and practitioners can benefit from the insights provided by this study when addressing the challenges faced by rural policing in China and other countries with similar circumstances.

Seasonal characteristics of crime: an empirical investigation of the temporal fluctuation of the different types of crime in London

Most types of crimes show seasonal fluctuations but the difference and similarity of the periodicity between different crimes are understudied. Interpreting the seasonality of different crime types and formulating clusters of crimes that share similar seasonal characteristics would help identify the common underlying factors and revise the patterns of patrolling and monitoring to enable sustained management of the control strategies. This study proposes a new methodological framework for measuring similarities and differences in the timing of peaks and troughs, as well as the waveforms of different crimes. The method combines a Poisson state-space model with cluster analysis and multi-dimensional scaling. A case study using twelve types of crimes in London (2013–2020) demonstrated that the amplitude of the seasonal fluctuation identified by this method explained 95.2% of the similarity in their waveforms, while the timing of the peaks covered 87.5% of the variance in their seasonal fluctuation. The high predictability of the seasonal patterns of crimes as well as the stable categorisation of crimes with similar seasonal characteristics enable sustainable and measured planning of police resource allocation and, thereby, facilitates a more efficient management of the urban environment.

Locating offenders: Introducing the reverse spatial patterning approach

ObjectivesCurrent strategies for locating where offenders live either focus exclusively on individual suspects or generalize to entire neighborhoods. However, better estimates of where offenders are located may improve models of the ecological distribution of crime, and forecasts of the locations of future crime incidents. MethodsWe propose a novel reverse spatial patterning (RSP) strategy that estimates where offenders may live based on the spatial locations of crime events. We rely on a distance decay function – based on the consistent finding that offenders do not travel far to commit crime – and Hipp's (2016) general theory of spatial crime patterns, to work backwards from the locations of actual crime events to make predictions about where offenders may live in subsequent years. We then use these estimates in models predicting crime locations. We create two versions of the RSP: one which assumes everyone is equally likely to offend, and another that creates an estimate assuming disproportionate offending across persons. ResultsWe test the effectiveness of our proposed strategy for these two measures using offense and arrest data from St. Petersburg, FL, and assess how well they predict the location of offenders (proxied by arrestees) and future crime events. We find consistent evidence that our RSP strategy provides better predictions of the locations of where offenders are located and also future crime incidents across a variety of crime types compared to existing strategies. ConclusionThe RSP approach is useful for creating estimates of where offenders live, which allow for better predictions of the locations of future crime incidents. These better forecasts will allow for more efficient allocation of police resources and targeted crime suppression efforts.

Early Detection of Suspicious Behaviors for Safe Residence from Movement Trajectory Data

Early detection of people’s suspicious behaviors can aid in the prevention of crimes and make the community safer. Existing methods that are focused on identifying abnormal behaviors from video surveillance that are based on computer vision, which are more suitable for detecting ongoing behaviors. While criminals intend to avoid abnormal behaviors under surveillance, their suspicious behaviors prior to crimes will be unconsciously reflected in the trajectories. Herein, we characterize several suspicious behaviors from unusual movement patterns, unusual behaviors, and unusual gatherings of people, and analyze their features that are hidden in the trajectory data. Meanwhile, the algorithms for suspicious behavior detection are proposed based on the main features of the corresponding behavior, which employ spatiotemporal clustering, semantic annotation, outlier detection, and other methods. A practical trajectory dataset (i.e., TucityLife) containing more than 1000 suspicious behaviors was collected, and experiments were conducted to verify the effectiveness of the proposed method. The results indicate that the proposed method for suspicious behavior detection has a recall of 93.5% and a precision of 87.6%, demonstrating its excellent performance in identifying the possible offenders and potential target places. The proposed methods are valuable for preventing city crime and supporting the appropriate allocation of police resources.

Seasonal mobility of populations and allocation of police resources in Finland

This study analyses the impact of the seasonal mobility of populations on the emergency response times (ERT) of the police. The concept of multi-locality increases the mobility of people but, however, the public service structures of the society and different administrative systems do not sufficiently account for that. The results of the mixed-model regression demonstrate the challenges in the police resources allocation when people are acting in different areas than expected. Emergency response times was lengthened as the seasonal population increased and spatially dispersed over the regions during the seasons. We suggest that the seasonal mobility should be involved the future planning process of police resource allocation.

An Integrated Graph Model for Spatial–Temporal Urban Crime Prediction Based on Attention Mechanism

Crime issues have been attracting widespread attention from citizens and managers of cities due to their unexpected and massive consequences. As an effective technique to prevent and control urban crimes, the data-driven spatial–temporal crime prediction can provide reasonable estimations associated with the crime hotspot. It thus contributes to the decision making of relevant departments under limited resources, as well as promotes civilized urban development. However, the deficient performance in the aspect of the daily spatial–temporal crime prediction at the urban-district-scale needs to be further resolved, which serves as a critical role in police resource allocation. In order to establish a practical and effective daily crime prediction framework at an urban police-district-scale, an “online” integrated graph model is proposed. A residual neural network (ResNet), graph convolutional network (GCN), and long short-term memory (LSTM) are integrated with an attention mechanism in the proposed model to extract and fuse the spatial–temporal features, topological graphs, and external features. Then, the “online” integrated graph model is validated by daily theft and assault data within 22 police districts in the city of Chicago, US from 1 January 2015 to 7 January 2020. Additionally, several widely used baseline models, including autoregressive integrated moving average (ARIMA), ridge regression, support vector regression (SVR), random forest, extreme gradient boosting (XGBoost), LSTM, convolutional neural network (CNN), and Conv-LSTM models, are compared with the proposed model from a quantitative point of view by using the same dataset. The results show that the predicted spatial–temporal patterns by the proposed model are close to the observations. Moreover, the integrated graph model performs more accurately since it has lower average values of the mean absolute error (MAE) and root mean square error (RMSE) than the other eight models. Therefore, the proposed model has great potential in supporting the decision making for the police in the fields of patrolling and investigation, as well as resource allocation.

The impact of COVID-19 lockdown restrictions on missing person reports

ABSTRACT The COVID-19 pandemic and the resultant restrictions imposed by governments pose short- and long-term challenges for the police, especially within resource-intensive areas of policing such as missing persons. However, the novelty of the pandemic means little research focus has been directed at understanding these challenges and how they may be overcome. Using archival reports from six UK police forces, the current study examined the extent to which missing persons cases changed during the first UK COVID-19 lockdown. Using a non-experimental fixed design, differences in the characteristics of reports of both children and adults who went missing between March and May 2020 and the same time period in 2019 were examined. Findings suggest a substantial reduction in missing reports overall, but a shift in the proportions of types of cases reported. For example, there was a greater proportion of adults classified as high risk of harm during this period and a greater proportion of children who were deemed low risk, who were living in residential care, and who were not from a White British background. Although forces must consider the findings within their local context, the study has implications in terms of demand and allocation of police resources, as well as multi-agency working. Future research is discussed.

Poverty as a Ground of Indirect Discrimination in the Allocation of Police Resources – A Discussion of Social Justice Coalition v Minister of Police 2019 4 SA 82 (WCC)

The Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000 (PEPUDA) prohibits indirect and direct unfair discrimination in terms of the grounds listed in the act (such as race, sex, and sexual orientation) as well as unlisted grounds (which are to be alleged and proven by an applicant). South African courts had also grappled with the specific issue of indirect unfair discrimination prior to the enactment of PEPUDA, where applicants could rely on the Constitution of the Republic of South Africa, 1996 directly. This is evident in cases such as Pretoria City Council v Walker 1998 2 SA 363 (CC) and S v Jordan 2002 6 SA 642 (CC). This contribution is an analysis of the pioneering judgment in Social Justice Coalition v Minister of Police 2019 4 SA 82 (WCC) (SCJ case), wherein a South African court for the first time recognised poverty as a ground of indirect discrimination under PEPUDA. This conclusion flows from the court's finding, based on expert evidence that the formula used to allocate police resources in the Western Cape unfairly discriminates against poor and Black people in an indirect manner. The analysis of the SCJ judgment will take place against the backdrop of the antidiscrimination framework under PEPUDA and direct constitutional litigation that predates PEPUDA. The underlying theme of intersectionality will also be discussed, as it was apparent from a reading of the SCJ case that grounds of discrimination often intersect with one another and disproportionately affect certain groups of people.

Spatio-Temporal Prediction of Baltimore Crime Events Using CLSTM Neural Networks

Crime activity in many cities worldwide causes significant damages to the lives of victims and their surrounding communities. It is a public disorder problem, and big cities experience large amounts of crime events. Spatio-temporal prediction of crimes activity can help the cities to have a better allocation of police resources and surveillance. Deep learning techniques are considered efficient tools to predict future events analyzing the behavior of past ones; however, they are not usually applied to crime event prediction using a spatio-temporal approach. In this paper, a Convolutional Neural Network (CNN) together with a Long-Short Term Memory (LSTM) network (thus CLSTM-NN) are proposed to predict the presence of crime events over the city of Baltimore (USA). In particular, matrices of past crime events are used as input to a CLSTM-NN to predict the presence of at least one event in future days. The model is implemented on two types of events: “street robbery” and “larceny”. The proposed procedure is able to take into account spatial and temporal correlations present in the past data to improve future prediction. The prediction performance of the proposed neural network is assessed under a number of controlled plausible scenarios, using some standard metrics (Accuracy, AUC-ROC, and AUC-PR).

Big Data and Procedural Justice: Legitimizing Algorithms in the Criminal Justice System

Big data algorithms are becoming more common in the criminal justice system: they are used to provide more effective allocation of police resources, to notify police of potentially dangerous individuals at specific locations, to guide efforts to intervene with individuals before they engage in criminal activity, to advise judges making decisions about pre-trial detention, and to provide guidance to judges at sentencing. The debate about the propriety of using these algorithms has primarily focused on whether they are fair to the defendants they are evaluating. However, those in favor of using these tools in the criminal justice system must also consider another potential hurdle: whether these new tools will be accepted by the general public. For centuries, decisions in the criminal justice system have been made by individuals: police officers, judges, bureaucrats, and legislators. The general public is accustomed to these decision-makers and, for the most part, accepts their legitimacy. However, the public may not be as comfortable with the idea that critical criminal justice decisions are made or heavily influenced by computer programs instead of human beings. This question is a critical one for big data supporters: even if these tools can provide predictions that are more fair, efficient, and accurate than the clinical judgments made by human beings, they will never become widespread if they cannot gain public support.

Examining Police Presence, Tactics, and Engagement as Facilitators of Informal Social Control in High-Crime Areas

The systemic model of social disorganization advocates that communities can help regulate disorder and crime by exerting informal social control. This paper examines the avenues through which police might contribute to that social process. Building upon past research, we compare the amount of police presence observed by residents versus residents’ awareness of and satisfaction with police–community engagement and tactics used as mechanisms for promoting safety, confidence in police, and social cohesion, and ultimately informal social control. Results of a structural equation model using data from a panel community survey in 71 high-crime areas suggest that simply seeing the police is not enough. The results suggest that tactics police use in neighborhoods and most notably, how those strategies are interpreted by residents living there can be a conduit to facilitate social cohesion and informal social control. The results have implications for police resource allocation and best practices in crime hot spots.

Crime Statistics: Modeling Theft in Favour of Victims’ Choices

The number of reported property thefts has dropped steeply in many European countries over the last 15 years. One reason for this could be that people have become more honest, which would imply that fewer resources should be allocated to the police and to crime prevention measures. In this paper, we have elaborated upon some alternative explanatory factors behind the decrease in the number of reported crimes within a utility-maximizing model where both thefts and victims are behaving rationally. Increased time and travelling costs for reporting, economic growth and a lower rate of solving crime could all explain the development in the reported crime rate. Within this theoretical approach, the nominal crime rate could decrease, while the real crime rate either remains constant or even increases. Applying the number of reported thefts as the sole or main parameter for allocating resources for crime prevention measures may result in a sub-optimal resource allocation. Developments in the number of reported thefts must therefore be supplemented by other indicators to provide a better basis to ensure an optimal allocation of police resources to combat theft. Estimates of unreported crimes must also be included as part of the basis of these allocation decisions.

Protest and property crime: political use of police resources and the deterrence of crime

This article investigates the claim that the political use of police resources promotes crime. Using a panel of South Korean metropolitan areas, we show that (1) the reallocation of police resources toward the control of political protests reduces arrest rates for crime and (2) the resulting reduction in criminal arrests significantly increases the incidence of crime. Overall, the impact of the reallocation of police resources works mainly through tradeoffs with arrest rates. Our findings imply that it is not the size of the police per se, but the allocation of police resources toward crime control that deters crime.

Using Hotspots Algebra to analyze crime events and their urban context

Hotspots analysis is essential in the criminology field and quite important in decisions making for police agencies because it permits the enhancement of allocation of police resources for timely and adequate actions. There exist different techniques for analysis and generation of hotspots, limited by its difficulty to consider other urban and demographic factors that could be the cause of the emergence of these hotspots or their influence over other factors. On the other hand, it is also difficult and unusual to analyze hotspots in a temporal context. This paper seeks, through spatial-temporal operations based in hotspots, to go beyond of classical crime analysis with hotspots, by looking for the spatial influence of other spatial factors over and analyzing also their relationship in a temporal context.
 The paper initially presents and analyze the performance of various techniques for hotspot generation and determines that STAC technique from CrimeStat is the more suitable for the proposed objective. Then, it defines a hotspot algebra allowing the combined study of crime and spatial factors affecting it and/or been affected by crime events. Temporal analysis includes the hotspots generation for days of the week and/or months in a year. In this way it is possible to study causality relationships and/or correlations among the studied phenomena and the spatial-temporal environment where crime occurs. Therefore, it is possible to define and apply informed actions, primarily concerning the allocation of police resources.
 Finally, it shows some application examples, thematic as well as temporal analysis, of hotspot algebra using crime data from Bogota for the years 2011 to 2013; finally, some future works in the subject are proposed.

The Influence of crime on active & sustainable travel: New geo-statistical methods and theories for understanding crime and mode choice

Abstract While several studies investigate the links between crime and travel behavior, little research examines the influence different types of crime or their location may have on active and sustainable travel choices. Furthermore, while travel is an inherently linear activity, most studies use coarse, zonal averages, likely missing key details important to pedestrians and bicyclists. In response, this study applies new geospatial measures and statistical methods to explore a number of interesting relationships between different crime types, their location, and their varying influence on green and healthy transit station access mode choices. Building on their previous work, the researchers uncover a complex set of relationships. Accordingly, this study suggests a new transportation research phenomena—the distinction between ‘property’ and ‘exposed’ modes. For example, property crimes appear to deter people from placing personal property at risk, such as parking a vehicle or a bicycle at a station, while violent crimes appear to significantly deter people from choosing to walk or bicycle, potentially exposing themselves to threats of personal injury. Our findings suggest that providing greater security for property modes at destinations (e.g. secure bicycle parking at stations), and personal safety for exposed modes (e.g. Crime Prevention through Environmental Design), may provide greater environmental and public health benefits than previously realized. This research infers important links between crime, policing, public health, and sustainability, informing a range of policy decisions, from urban design to police resource allocation. As most public policy and design strategies can take years to show results, this research helps guide crime intervention policies that may more quickly achieve important, health and sustainability goals.

K-Shortest-Path-Based Evacuation Routing with Police Resource Allocation in City Transportation Networks.

Emergency evacuation aims to transport people from dangerous places to safe shelters as quickly as possible. Police play an important role in the evacuation process, as they can handle traffic accidents immediately and help people move smoothly on roads. This paper investigates an evacuation routing problem that involves police resource allocation. We propose a novel k-th-shortest-path-based technique that uses explicit congestion control to optimize evacuation routing and police resource allocation. A nonlinear mixed-integer programming model is presented to formulate the problem. The model’s objective is to minimize the overall evacuation clearance time. Two algorithms are given to solve the problem. The first one linearizes the original model and solves the linearized problem with CPLEX. The second one is a heuristic algorithm that uses a police resource utilization efficiency index to directly solve the original model. This police resource utilization efficiency index significantly aids in the evaluation of road links from an evacuation throughput perspective. The proposed algorithms are tested with a number of examples based on real data from cities of different sizes. The computational results show that the police resource utilization efficiency index is very helpful in finding near-optimal solutions. Additionally, comparing the performance of the heuristic algorithm and the linearization method by using randomly generated examples indicates that the efficiency of the heuristic algorithm is superior.

Complaints of group-stalking (‘gang-stalking’): an exploratory study of their nature and impact on complainants

Stalking primarily concerns the actions of individuals. However, some victims report stalking by organised groups, this being known as ‘group-’ or ‘gang-stalking’. This phenomenon has not been subject to systematic study. An anonymous questionnaire was completed online by self-defined victims of stalking. One thousand and forty respondents met research definitions for stalking, of which 128 (12.3%) reported group-stalking. One hundred and twenty-eight individually stalked cases were randomly selected as a comparison group. All cases of reported group-stalking were found likely to be delusional, compared with 3.9% of individually stalked cases. There were highly significant differences between the two groups on most parameters examined. The group-stalked scored more highly on depressive symptoms, post-traumatic symptomatology and adverse impact on social and occupational functioning. Group-stalking appears to be delusional in basis, but complainants suffer marked psychological and practical sequelae. This is important in assessment of risk in stalking cases, early referral to psychiatric services and allocation of police resources.

A105 The Meaning of “Mean Streets” for Sustainable & Active Travel: Crime, (Pedestrian/Bike) Casualties and Mode Choice

While travel is an inherently linear activity, most studies of the influence of the urban environment on active travel oftentimes attribute coarse, zonal averages to unique individuals, likely missing key details important to pedestrians and bicyclists. This research fills a key knowledge gap by examining: 1) The influence of different types of criminal activity (e.g., property vs violent crime) along different segments of a person’s travel route; 2) The influence of bicycle and pedestrian collision data in combination with criminal activity on sustainable and active travel choices. Finally, this is one first studies to align urban environmental quality measures along an individual’s path, from origin to destination, as well as provide a detailed examination of the choice to bicycle in lieu of other, available modes. This study employs unique methods and metrics through an innovative application of individualized linear spatial units of analyses in combination with detailed geospatial data (on crime, pedestrian and bicycle collisions, and urban design factors) - in sum, disaggregated data for disaggregated research. Many unique datasets were gathered and innovatively applied in this study, including: Geo-located point data on criminal activity, categorized as per the U.S. Department of Justice (major and minor property crimes, major and minor violent crimes, and instances of vandalism), and tested for their varying influence around the respondents’ home, route, and destination. Geo-located point data on pedestrian and bicycle collisions along the respondents’ routes. Many other data on the built environment, socio-demographics, and other important factors (parking availability). Finally, this study analyses these data through various multinomial logit (MNL) predictive models of the choices to walk, bicycle, ride bus, drive-alone, or be dropped-off. The predictive, MNL model results find that different crimes appear to have different effects on different modes: property crimes deter people who may wish to avoid placing personal property at risk—such as parking a vehicle or a bicycle. Violent crimes along-the-route appear to have a significant deterrent to modes where travelers are more exposed to personal risk, such as walking, bicycling and transit ridership. People who may have the option to avoid certain stations with high threats to personal safety, appear to exercise that option, such as people being dropped-off, carpooling or driving alone. Finally, when pedestrian and bicycle casualties are entered into the model, there are changes to the associations with criminal activity. By improving our understanding of the connection between the experiential quality of street ecology (as influenced by crime, pedestrian/bike casualties, and urban design) on people’s choice to engage in sustainable and active travel, This study provides important policy guidance on such efforts as creating safer, more livable streets, CPTED, and the allocation of police resources.