Fusion Centers: Evolving Roles, Expanding Capabilities, Continuing Challenges
By Juan Colon, John Gillon, Gretchen Stewart, Robert Morison, Aug 24, 2020
Fusion centers have evolved far beyond their original role in counter-terrorism to safeguarding and serving the public on many fronts. They continue to evolve rapidly as they opportunistically and, in the COVID-19 pandemic, necessarily make their operations more virtual. The core challenges are also changing, but they still revolve around the three imperatives of integrating data, deploying technology, and building trust. Three experts – Juan Colon, National Director of Opioids and Illegal Drug Solutions, SAS Institute, John Gillon, Industry Expert, Public Security, SAS Institute, and Gretchen Stewart, Public Sector Data Science Director, Intel Corporation – share their experience and perspectives on how fusion centers succeed in their missions today and tomorrow.
Fusion Centers Then and Now
Fusion centers are information and intelligence hubs designed to help governmental agencies share knowledge and expertise about crimes, threats, and hazards. They are collaborative information sharing environments that facilitate the gathering and analyzing of data from many sources and turning it into useful intelligence. The results enable agencies to anticipate threats, prevent crimes, respond to emergencies, inform policy development, and deploy resources more effectively.
Fusion centers were born after the 9/11 terrorist attacks, when the 9/11 Commission concluded that there had been not only failures to share information, but also poor intelligence analysis and “failure of imagination.” While fusion centers have evolved, recent terrorist tactics (primarily against European civilian targets) have varied from sophisticated, highly coordinated attacks to crude blunt-force strikes. These have also revealed failings in the local, national, and international intelligence and enforcement services. They are perceived to have missed opportunities to preempt such attacks, which have been largely perpetrated by known (often petty) criminals or “lone wolves” who had become radicalized. The growing need to recognize the telltale behaviors and actions of these local actors is blurring the lines between traditional law enforcement and counter-terrorist activity and has elevated general policing and community information to among the most critical data sources for fusion centers.
Today, there are hundreds of fusion centers around the world and 80 in the U.S. alone. Fusion centers operate under four critical operational capabilities (COCs), which are defined as: Receive, Analyze, Disseminate, and Gather. However, no two fusion centers are quite alike. They have different data collection and analytical capabilities. They may serve major cities, regions, states, nations, or beyond. The Europol center combats terrorism and crime across the European Union. In the U.S., fusion centers have different partner agencies, including federal agencies such as the Federal Bureau of Investigation, Drug Enforcement Administration, and Department of Homeland Security. They serve as a conduit of information between the 16 agencies within the Intelligence Community and the state and local police departments.
Most fusion centers have evolved and expanded beyond their original counter-terrorism missions, and may have other focus areas such as organized crime, gang activity, or cybercrime. Most support local law enforcement and may also support the various emergency management functions related to sporting, cultural, and political events; natural and man-made disasters; and public health problems, including the opioid crisis and the COVID-19 pandemic. Additionally, fusion centers serve a variety of private sector constituencies within national critical infrastructure sectors, including energy, manufacturing, transportation, health, and communications.
Overall, fusion centers play a vital role in supporting security and protection efforts ranging from everyday law enforcement to local and national crisis response. They prove essential whenever cross-organization intelligence sharing is critical to public safety and national security.
Despite the great variety of fusion centers, they all have three common key ingredients or success factors – data, technology, and trust. As fusion centers have scaled and morphed to take on expanded responsibilities and address additional public safety needs, these continue to be the success factors. We’ll explore how they come together in a high-performance fusion center, but first let’s look at how a couple of centers operate.
New Jersey Regional Operations & Intelligence Center
Launched in 2005, the fusion center is the state’s primary focal point for information sharing and intelligence production to support law enforcement, counter terrorism, homeland security, and other public safety agencies. Here are examples of the center in action.
In April 2013, when the Boston Marathon bombing occurred, the area’s fusion center went into overdrive – gathering, analyzing, and sharing data in real time to identify the bombers, prevent any subsequent acts, and enable spectators to safely depart the area. Through the Homeland Security Information Network, the New Jersey fusion center, like those in other states, monitored the evolving situation. Leveraging limited information about the suspected bombers, the center uncovered that the bombers had relatives residing in New Jersey. This prompted a response by federal agencies to the residence in the event that the bombers, who were on the run, were fleeing there.
When Hurricane Sandy struck in October 2012, the New Jersey fusion center served as the emergency operations center by accommodating the personnel coordinating the 15 emergency management essential support functions. More than 100 personnel operated out of the fusion center to facilitate communication, information sharing, and coordination of response efforts. As some local first responders’ facilities and equipment were decimated, the fusion center facilitated the deployment of various resources to support or supplant the affected local first responders, who had been overwhelmed. As a result, they facilitated in rescuing people, preventing looting, and directing people to emergency shelters to obtain food and fuel.
Around 2012, when governments started to widely recognize the impact of the opioid epidemic, the fusion center had already accumulated several years of drug incident data, which immediately enabled law enforcement and public health agencies to understand the presence and prevalence of drugs and their impact on citizens. The fusion center analyzes various data sets enabling both law enforcement and public health services to deploy resources more effectively to the areas of the state that are most severely impacted by drugs.
Scottish Crime Campus
An example of close collaboration and changing operational mandate, the Scottish Crime Campus opened in 2014. It is a comprehensive national fusion center with an unusually integrated structure and outcome focused operational model. Twenty different agencies work within the Campus, including the police, customs and excise, forensic services, the national crime agency, and the fire service. But it’s difficult to tell who is with what agency. The identification badges don’t say. Personnel receive common training in areas like intelligence analysis, interview techniques, and crime scene preservation, and they work to the same standards.
This “all in it together” structure and culture makes it much easier to share information and see results. Teams can, for example, paint more complete pictures of criminal activity by drawing data from police surveillance and investigation, folding in tax and revenue data, and taking advantage of technical intelligence gathering capabilities; something that, in the normal mode of operations, only a few of the partner agencies would possess.
With all agencies “in house,” the Scottish Crime Campus has more direct responsibility for execution than typical fusion centers do. Each case is led by the agency most likely to meet the objective, often to secure a successful prosecution. Campus staff don’t just pass information and analysis on to local authorities – they follow through as part of the case teams.
Comprehensive capability and close cooperation make for better intelligence, smarter execution, and consistently high rates of apprehension and conviction of criminals. Working across borders, the Campus also participated in the capture of several hundred international “most wanted” criminals in its first two years of operation.
The objectives and ambitions of fusion centers present enormous challenges in the data domain. They need data that is as accurate and complete as reasonably possible, data that is timely and increasingly real-time, data that is integrated enough to describe complex situations and provide rich data sets for analytics. Driven by the proliferation of connected devices and social media, data volumes are already enormous and growing fast. There are many types of data, both structured and unstructured, including heavy reliance on textual data and increasing reliance on voice, video, and geo-spatial data. And the data comes from many sources – not just from different agencies, but from disparate information systems within individual agencies. Ultimately, data aggregation and curation pose greater challenges for fusion centers than the sheer volume of data does.
Fusion centers may have personnel from local, state, and national agencies working together and sharing information, but that doesn’t mean the data they use is integrated. Individual agencies still struggle to make their data good enough to share. Their data needs to be labeled and able to be extracted, transformed, and loaded into databases and analytical models that can be trained for accuracy and optimized for performance.
One of the specific challenges is resolving entities – people, objects, locations, and events – that are represented differently in databases within and across agencies. The center needs, for example, to find all the information about any given person, vehicle, or location; make sure that the information is indeed about the same entity; and then find the connections with other well-identified entities. All data must be secured, access-restricted as required, and processed in compliance with any sharing caveats or legislative imperatives. And if it is personal protected data, such as health-related information about an individual, it must be managed and used in ways that comply with privacy laws and stand up to ethical scrutiny. Some of the data containing personal information must be anonymized to make it appropriate for use.
The bottom line is that fusion center data is very difficult to manage and extremely difficult to integrate for effective use. Too much potentially valuable data – especially unstructured data – cannot be exploited fully, so patterns and predictable behaviors go undetected.
If we look across the entire data management process – data acquisition, data aggregation and curation, model development and scoring, and continuously updating with new data – the fusion center is in a perfect storm. Data management is essential, but new methods are needed. If much of the data is at the ready in well-organized databases, so much the better. But for the most part, the methods of conventional master data management don’t serve in fusion centers. The processes of cleansing and structuring data are too slow. Data needs to be gathered, integrated, analyzed, and used faster and more opportunistically. That brings us to the second success factor.
Given that perfect storm, fusion center data management processes can’t be slow or manual, and they can’t afford to consume valuable analyst time with data entry and cleansing tasks. The good news is that more advanced data management and analytics technologies, increasingly incorporating AI, address the issues of data aggregation and curation, along with model deployment and scoring.
However, leveraging the latest and best technology requires a sound architecture to begin with, and partner agencies typically have fragmented technology architectures alongside their fragmented data stores. They often make matters worse when they try to incorporate new technologies through complicated and hard-to-maintain interfaces. Over time, piecemeal technology interferes with data integration, analysis, and sharing.
Fusion centers need to avoid those problems by implementing consistent and manageable platforms to enable their work. The key word is “platform,” a robust, integrated, and flexible core of computer processing, communications, and applications. How is a technology platform architected, and what can it do? It all starts with a well-governed data strategy.
The core platform can serve the large majority of everyday tasks and workflows in efficient, consistent, and predictable fashion.
That means it must be able to ingest then transform huge amounts of data from a variety of sources, from crime data and statistics to real-time sensor feeds.
The platform enables fusion center professionals to serve themselves by using inquiry, analysis, scenario, and visualization tools, putting their domain expertise to work without having to become technology experts. They become “citizen data scientists.”
The platform is open to new data sources and analytical methods, including open-source products and outputs from specialized technologies like AI engines.
The platform connects with wherever the data comes from and needs to go to, including edge computing and web environments through the cloud.
Perhaps most important for the investigative work of the fusion center, the platform has capabilities to handle large amounts of streaming data and distinguish entities and their connections on the fly. Data examination and analysis can be continuous in real time.
A robust platform isn’t one monolithic system. It could be hybrid – systems in house, bursting and leveraging other data stores from cloud providers, research institutes, and global organizations. The platform’s capacity and capabilities evolve with the mission and requirements of the fusion center. And it serves the core operations of the center every step of the way.
In essence, a fusion center is a means of enabling and encouraging communication, cooperation, and coordination in anticipating and responding to threats to the public. But there are organizational, political, and sometimes legal barriers in the way when partner agencies have different structures, working practices, priorities, and cultures. Where their spheres of operations overlap, they may be rivals. And even when they are motivated to share information, regulations governing information access and use can prevent it. Overcoming these barriers starts with the commitment to cooperate among the leaders of the agencies, and sometimes their communal effort to influence legislation.
The biggest barrier, however, may be at the grassroots level. It’s the matter of trust. Sometimes there’s active distrust resulting from a lack of familiarity with a fusion center’s operations and purpose, or it may result from past disagreements between agencies over investigations. Often these were impediments for fusion centers when they were being established. Some local police departments are hesitant to share their data because they’re unaware or uncertain about how the fusion centers would use it. Each agency needs to be confident that its information will be handled properly, kept secure, and used for mutual benefit. And sometimes agencies don’t trust their own data to begin with. The risks of not sharing may seem lower than the risks of sharing incomplete or inaccurate information.
How do fusion centers build the necessary trust? One direct way is co-location, as at the Scottish Crime Campus. Have representatives of partnering agencies work in the center and learn to collaborate and trust each other day to day. With or without co-location, a fusion center needs to know its “customers” and the primary problems they seek to address. And it needs to educate those partnering agencies on several fronts:
What the fusion center is, its purpose and functions, and its capabilities and limitations.
The agencies’ roles in the fusion center processes and how to perform them well, especially their responsibilities for generating and contributing data as well as receiving and acting on it. They may need to appreciate the value of their data and the value of investing in it.
The mutual value proposition, which is by analogy, “Give us 25 cents worth of data, and the center will give you a dollar’s worth back.”
It may take time for fusion center partners to appreciate the value that is being generated. In the U.S., a healthcare agency initially refused to share any data with the fusion center because of privacy concerns. The fusion center shared some of its analytical products with the agency anyway. In time, the healthcare agency realized the value being produced, saw that data was being managed and used responsibly, and commenced formal data sharing with the fusion center.
Fusion centers may also need to earn the public trust and that of advocacy organizations like the American Civil Liberties Union (ACLU). That requires transparency regarding the center’s privacy, civil rights, and civil liberties policies and how they are developed. In short, for all the challenges fusion centers face with data and technology, trust can sometimes be the hardest element to put in place. But without trust, fusion centers may limit their access to the data they need to integrate and analyze to enhance the safety of the public.
Putting the Pieces Together
With ample data, technology, and trust, the high-performing fusion center can develop intelligence and act on threats quickly and effectively. That’s because it can excel operationally across the entire intelligence life cycle:
Planning – setting the specific objectives that drive the investigation and the priorities along the way.
Collection – determining what data from many sources is relevant and available, or needs to be generated, and then gathering it together.
Processing – manipulating and integrating the data to be useful for inquiry and analytics, while abiding by any restrictions on data use.
Analysis – using statistical and advanced analytical models, along with rapid deployment methods, to turn data into insight, including developing and testing hypotheses and scenarios.
Dissemination – creating and distributing specific products, like risk assessments and target profiles, for specific constituencies and decision-makers.
Evaluation – providing appropriate feedback to determine if the needs of the customers are being met with appropriate analysis and relevant products.
The life cycle is completed as the fusion center continuously monitors and improves data provisioning, model performance, intelligence products usage, and overall strategy, process, and governance.
The high-performing fusion center can excel across two distinct levels of analysis. Criminal intelligence is tactical. It seeks to support operational activity and inform decision makers by looking at individuals, their patterns of crime, and their criminal networks. Who are the criminals? Where are they operating? What crimes have they been involved in? Who are their associates? What are their capabilities, and what potential threats do they pose? Descriptive analytics and social network analyses provide the answers.
Crime pattern analysis is more strategic. It takes a more macro-level view across an area, a city, maybe statewide or even national. What are the patterns and trends in criminal activity? What are the hotspots for specific types of crime? What anomalies stand out and possibly indicate that patterns are changing? What are the opportunities to intervene? Where and when can law enforcement and other resources be deployed to greatest effect? What law enforcement capabilities need to be expanded or improved? Forecasting techniques, anomaly detection, and ensembles of predictive and prescriptive analytics models provide the answers. And then the loop is closed when strategic data informs local criminal intelligence.
As mentioned, fusion centers and their missions have expanded greatly from their initial charters around homeland security and counter-terrorism. First to addressing local crime issues, then supporting emergency management functions, and in recent years, addressing public health matters. The centers’ roles in these areas continue to expand, and new ways may emerge for repurposing their core information sharing and intelligence generating capabilities.
The race to leverage more data better will continue. We say “race” because the pace continues to increase as data sources and volumes continuously expand. Fusion centers will be incorporating more external data, not only about demographics and public infrastructure, but also other research data, including social science, gender, and ethnicity data from sources like the Dataverse Project. There will be similar ongoing challenges in the technology realm, as fusion centers take advantage of emerging technologies while striving to keep their platforms and data strategies coherent, flexible, and scalable. And as fusion centers have brought crime analytics to the fore, predictive analytics capabilities and applications will continue to expand.
Following the 9/11 Commission’s recommendations, fusion centers were established as bricks-and-mortar locations. Having representatives of different agencies co-located has proven invaluable, as at the Scottish Crime Campus, especially when it helps compensate for less-than-integrated data and technology. Co-location breaks down cultural barriers and facilitates multi-agency collaboration through personal networking, sharing of resources, and joint operations.
However, the recent trends toward automating processes and reducing operating costs, and now the rapid shift to remote work amid the COVID-19 pandemic, are accelerating the move to virtual fusion centers. With many fusion center staff working from home, the pandemic has given centers some forced practice, and some practical lessons learned, in how to operate and collaborate remotely. The technology is there to operate virtually, and for fiscal and logistical reasons, centers are establishing virtual pathways forward. The interim step for many will be hybrid bricks-and-mortar and virtual centers. In the transition, the challenge will be to reproduce the proven benefits of co-location and to continue building and maintaining trust in new ways.
Finally, local law enforcement agencies under increasing fiscal constraints are going to need the services of fusion centers more than ever to provide them intelligence and supplement their capability.
Checklist for Action
We summarize and conclude with recommendations for fusion center and agency leaders to help them maximize the performance and value of their fusion centers.
Know your partner agencies and other customers and develop information and intelligence products to meet their specific requirements and priorities.
Educate your partner agencies and other customers as the capabilities and products of the fusion center evolve.
Drive everything with the end goals in mind. Everyone, from expert investigators to the technologists managing the platform, should operate from a specific set of desired outcomes.
Stay operationally focused. The fusion center is a central information hub whose objective is to put information and actionable intelligence in the hands of people on the front lines.
Be prepared for spikes of new data and demand for intelligence, both in times of emergency and when the center starts working with new partners.
Expand the range and diversity of both data sources and partners, such as industry experts and community groups. Valuable data can materialize in many places.
Maintain close relationships with leaders of partnering agencies, get strategic input from them, and when needed enlist them in advocating for legislative change.
Keep leveraging technology to automate fusion center processes. That’s the only way to keep up as data continues to grow exponentially.
Keep innovating in those processes, including by exploiting emerging technologies.
Keep building trust and nurturing a culture of collaboration within the fusion center and across its partner agencies.
About the authors
Juan Colon serves as an advisory industry consultant responsible for enabling the development of technologies around the opioid crisis and public safety and homeland security matters for the U.S. government and Latin America.
He retired at the rank of Major after 25 years from the New Jersey State Police, where he served in several diverse roles to include operating in an undercover capacity, implementing and operating New Jersey’s fusion center, developing and implementing statewide data projects, and serving as a drug policy advisor at the New Jersey Attorney General’s Office. He developed the Drug Monitoring Initiative (DMI), which was included in the National Heroin Strategy report by the Office of National Drug Control Policy (ONDCP), as a recommended strategy to confront the opioid epidemic.
The Office of the U.S. Attorney General, the Bureau of Justice Assistance, the Centers for Disease Control, ONDCP, the National Governors Association, and the National Security Council have leveraged his expertise in this domain to present at national level conferences and panel discussions.
John Gillon joined SAS as a Consultant having retired as a Detective Superintendent in Scotland’s then largest force, Strathclyde Police.
In a police career spanning 30 years, John developed a professional profile encompassing many aspects of policing with emphasis on the expert areas of investigation and intelligence. His service incorporated many specialist roles in which he accrued extensive experience in the investigation of serious and organised crime and national security matters.
John’s rich and varied career includes the position of Programme Manager for a national change initiative to deliver common integrated information and communication systems. Since joining SAS, John has continued to build on his expertise, consulting globally with many policing agencies embarking on modernisation programmes.
Gretchen Stewart is the Public Sector Data Science Technical Director working closely with Intel’s Ecosystem of software, hardware and partner vendors in developing solutions and programs focused on the convergence of high-performance computing and data analytics. In building this data analytics/AI practice within Intel she is hands on with federal and state government, research institutions, universities, energy, and advanced manufacturing customers. In enhancing Intel’s ecosystem with these AI solutions and programs, she is working with disruptors, start-up and mainstream software vendors to ensure governments, educational institutions, and enterprise customers advance the future of research, have competitive advantages, build smarter cities, and provide enhanced services to citizens.
She holds a bachelor’s degree in mathematics from Wells College and an MBA degree from University of New Hampshire System, along with leadership and branding certifications from Harvard and Dartmouth. She completed a nine-month data analytics certification jointly developed by Harvard Business School, Harvard School of Engineering and Applied Sciences, and Harvard Faculty of Arts and Science in June 2019.
Robert Morison, Lead Faculty Member, IIA
Robert Morison serves as Lead Faculty with IIA. He is an accomplished researcher, writer, speaker, and management consultant, and an authority on what happens at the intersections of business, technology, and people management. He has been leading breakthrough research for more than 30 years, collaborating with eminent academics, thought leaders, and management innovators. He has written on topics ranging from business innovation, reengineering, and analytics to workforce management, demographics, and retirement.
Bob is coauthor of three books: What Retirees Want: A Holistic View of Life’s Third Age (Wiley, 2020), Analytics at Work: Smarter Decisions, Better Results (Harvard Business Press, 2010), and Workforce Crisis: How to Beat the Coming Shortage of Skills and Talent (Harvard Business Press, 2006). His 2004 Harvard Business Review article, “It’s Time to Retire Retirement,” coauthored with Ken Dychtwald and Tamara Erickson, received a McKinsey Award. He holds an AB from Dartmouth College and an MA from Boston University.