All posts by Guest Author

Current Operations

Why the Sudden Drop in Armed Robbery of Ships off Venezuela?

1 Comment

By Lydelle Joubert

Venezuela has the largest known oil reserves in the world, with 302 billion barrels of proven reserves reported in 2018. Production has however declined from 3.5 million barrels a day in 1997 to 1.4 million barrels in May 2018 due to the persistent economic crisis the country has been facing.

Anzoátegui state is traditionally one of the largest oil production centers in Venezuela. Two thirds of Venezuelan oil is exported through Puerto Jose in Anzoátegui. The concentration of oil industry-related infrastructure in Anzoátegui combined with the declining economic situation and lack of security make it a hotspot for armed robbery at sea in the Caribbean region. Several anchorages lie off the coast of Anzoátegui, such as Bahia De Barcelona Anchorage, Puerto Jose Anchorage and Puerto La Cruz Anchorage. Due to the collapse of the fishing industry, the economic hardship that coastal communities are facing and insufficient security measures at these anchorages, men in small boats approach mostly tankers waiting to load oil and board vessels in order to rob them.

According to the definition of the United Nations Convention on the Law of the Sea (UNCLOS, 1982: 61) piracy is limited to acts outside the jurisdiction of the coastal waters of a state. Acts committed in territorial waters are considered armed robbery of ships. As these anchorages are located in Venezuelan territorial waters, it is classified as armed robbery of ships.

For unknown reasons, reports of armed robbery at these anchorages stopped in the middle of 2019. Between January 2016 and the end of April 2019, 36 robberies and attempted robberies were reported at anchorages off Anzoátegui, of which 29 were on tankers. Six incidents were reported in early 2019, but in April 2019 robberies on commercial ships at these anchorages ended abruptly. No robberies were reported for the next ten months, until the end of February 2020.

Robbery of Vessels at Anchorages off Anzoategui, Venezuela (via Stable Seas Database)

What happened in late April or early May 2019 to account for this change? Understanding the cause of this change is important for predicting whether this sharp fall in armed robbery is sustainable or likely to be reversed in the future. Before analyzing potential causes of this sharp decline in armed robberies, it is useful to review what happens in a typical armed robbery at sea in this area.

What Happens in an Armed Robbery at Sea?

Most robberies on ships at these anchorages can be classified as petty theft where a ship’s stores and its crew’s possessions are stolen. Three to seven robbers in small boats approach anchored vessels under darkness and board via the anchor chain and hawse pipe or via a grappling hook and rope. Robbers are usually armed with knives, but guns were observed in a few cases. In a few instances the crew was tied up, threatened, or assaulted and minor injuries were reported.

During a more brazen robbery on 14 October 2018 the bulk carrier Shi Zi Shan was boarded just after midnight by four armed men in national guard uniforms under the ruse of an anti-narcotics inspection. Once aboard they threatened the crew with handguns and commanded them to be taken to the captain’s cabin. They stole all the cash and crew’s valuables.

Most yachts have long since departed this coast, ever since the attacks on these ships turned violent when the Dutch captain of the yacht Mary Eliza was shot and killed at Marina El Concorde in September 2013. These incidents, combined with recent kidnappings of Trinidadian fishermen on the coast of Venezuela, created fear (although unfounded) that piracy and armed robbery off Venezuela could turn into a situation similar to Somalia where crew from commercial vessels are kidnapped from vessels for ransom.

What Contextual Changes Might Account for the Fall in Armed Robberies at Sea?

In January 2018 Venezuelan President Nicolas Maduro temporarily closed maritime borders with the ABC islands (Aruba, Bonaire, Curacao). Fresh fruit and vegetables from Venezuela were transported to these islands, but the same routes were also used to smuggle gold, silver, copper and coltan from Venezuela.  In February 2019 maritime borders with the ABC islands were once again closed, this time to prevent humanitarian aid from reaching Venezuela from these islands. The measure applied to commercial and fishing vessels. This led to a larger military presence in ports and an increase in vessel inspections on vessels entering ports in an effort to stem the smuggling of aid from the neighboring ABC islands.

At the same time, other economic factors may have reduced shipping traffic and could have reduced opportunities for armed robbery at Puerto Jose. Venezuela suffered from power outages that affected oil production and shipping operations. On 5 April 2019 the U.S. Treasury imposed sanctions on tankers and shipping companies transporting Venezuelan oil to Cuba, which could have further reduced arrivals of vessels at the anchorages. Tankers looking to evade the sanctions would have strong incentives to turn off transponders near Venezuelan water, making it difficult to obtain accurate counts of vessels at these anchorages. This could also dissuade reporting of armed robberies.

Armed robbery of ships at anchorages off the coast of Anzoátegui, 2016 to February 2020. (Stable Seas Database)

However, these explanations are not satisfactory because not all countries adhered to the call for sanctions and Venezuelan vessels were still operating from the port and anchorages. Armed robbers had ample opportunities to commit crimes near Puerto Jose. A closer look reveals that one specific incident may have initiated a chain of events that led to this decline in armed robberies.

How One Act of Defiance May Have Changed Incentives for Armed Robbery

There was one incident that fit the timeframe coinciding with the end of robberies on vessels at these anchorages, but the fact that it spelled a halt of these crimes was quite unintended.

It appears that dissatisfaction is growing amongst mid-level Venezuelan officials who are unhappy with the government for supplying Cuba with oil while severe shortages are experienced within Venezuela. In previous months, nationwide protests were reported at Venezuelan state-owned, Petróleos de Venezuela (PDVSA) facilities, leading to a loss of 30 percent in production. On 1 May 2019 a captain of a PDVSA tanker, Manuela Saenz, defied orders to deliver oil to Cuba and notified PDVSA of his intent. 

Members of the Bolivarian intelligence service (Sebin) allegedly boarded the products tanker near the Amuay terminal and took control of the vessel. The captain was replaced and Sebin members remained onboard while the delivery to Cuba was made. AIS was also switched off for most of the trip.

Since then, Venezuelan Armed Forces (FANB) have been deployed on 15 PDVSA-operated tankers to ensure that fuel is delivered to Cuba and that crew would not sabotage tankers or divert the product. It is also speculated that armed personnel are placed on the vessels in case the U.S. blocked shipments to Cuba.

This added armed security on these tankers is in all likelihood the determining factor why armed robberies off Anzoátegui stopped at the end of April 2019, but this lull in attacks was short-lived. On 24 February 2020, six armed men wearing balaclavas boarded the tanker San Ramon anchored near Isla Borracha, north of Puerto La Cruz notwithstanding the presence of a coast guard armed guard onboard. This time violence escalated. The captain, Herrera Orozco, resisted the robbers and was shot in the face and killed. Another crewmember is still missing after he jumped overboard, and a coast guard sergeant was injured during the attack.

Conclusion

While we cannot know what caused these armed robbers to be more violent than those involved in previous incidents, it is plausible that this is related to increased security on ships, and the situation escalated. Ships at these anchorages are harder targets than they once were, but the root causes of piracy and armed robbery in Venezuela – including poverty and weak state governance – remain. So long as they do, it is possible that attacks on ships at these anchorages will be dissuaded only as long as criminals can be deterred from employing escalating levels of violence.

Lydelle Joubert is an expert on maritime piracy at Stable Seas, a program of One Earth Future. She has an MA in International Relations from the University of Pretoria, South Africa.

Featured Image: Venezuelan military policewoman in a presidential meeting. (Wikimedia Commons)

Future Tech

Don’t Neglect the Easy Wins for Military AI

1 Comment

By Christian Heller

The defense community is captivated with artificial intelligence (AI) and its possible impacts on warfare. There has been much debate on AI’s impact on offensive and defensive operations, nuclear command and control, and information warfare. AI experts worry about the dangers of ultrafast AI decision-making, the U.S.-China AI arms race, the level of autonomy granted to robots, and the overall threat to humans with increased AI independence.

The Department of the Navy (DoN) has responded with building organizations to help integrate AI into the military services. Most prolific of these groups are the Algorithmic Warfare Cross-Functional Team (also known as Project Maven) and the Joint Artificial Intelligence Center (JAIC). Project Maven’s efforts focus on using AI to support the processing, exploitation, and dissemination (PED) of video and imagery intelligence. These efforts also include using AI to exploit captured enemy material (CEM), acoustic intelligence (ACINT), and publicly available information (PAI, also known as Open-Source Intelligence, or OSINT). The JAIC’s first two initiatives were predictive maintenance and humanitarian assistance/disaster relief. Later last year, they expanded to include cyberspace and robotic process automation.

While these lines of effort are important, they pursue difficult, hard-to-achieve tactical goals while ignoring the easy, low-hanging-fruits of AI implementation within the bureaucracy. Self-targeting drones, deep fakes, and global integrated predictive analytics platforms are worthwhile, but the Navy can achieve better and faster returns on its investments by pursuing unglamorous AI efforts in the fields of manpower and administration. With a renewed willingness to rebuild the services to face the threats of the future, now is the ideal team to embrace AI.

Manpower

Existing uses of AI in the private sector can be implemented by the DoN to support recruiting, training, retention, promotions, and billet assignments. Numerous companies are using AI to help their hiring managers identify and recruit employees. Recruiting commands could adopt these services to reduce their personnel burden and increase their effectiveness. Montage combines AI, process automation, and analytics to personalize the recruitment process toward specific candidates. Textio helps recruiters choose the right words and language to attract the right people. Firms like Koru use predictive AI to better match candidates to available positions and could change the way the services assign personnel to specialties and units.

Many companies have already implemented cost-saving AI measures such as these. Google worked with American Eagle to customize their marketing to individual consumers, similar to how the services could customize efforts for individual recruits and better manpower management. Amazon, Starbucks, and Nike all use AI to personalize customer engagement and marketing. LinkedIn uses AI for its LinkedIn Recruiter platform to identify the best candidates for hiring managers, and Home Depot and Dyson use AI programs to identify candidates based on their internal databases, social media, and public job boards.

The lack of continuity of knowledge is endemic to the DoN where service members continuously change billets and commands. Turnover is high and leads to a severe lack of institutional knowledge. This turnover means the time-cost of retraining a replacement detracts from time spent advancing a project forward. AI training systems can help. IBM has partnered with firms to help departing employees document their knowledge for future workers. AI then indexes and sorts the information to make it more easily available, and successful efforts have already reduced the search times for previous knowledge by 75 percent.

The Navy and DARPA already proved the relevance of AI to training service members. A combined project in which a digital AI tutor led new sailors though their training saw AI-trained students “frequently outperform Navy experts with 7-10 years of experience.” AI startups like Bakpax aid teachers with their grading to identify specific personal needs for students and speed up the corrective process. A study by Johns Hopkins University found that students using Knewton, one of the original AI education startups which personalizes learning plans and materials for students, performed better compared to peers. Improvements in training may not seem like a critical requirement for the services, but training and development is viewed as the primary job benefit by millennials in choosing their employer.

AI also can help reform promotion processes with are plagued with inefficiency. Analytics firms like Palatine are helping leaders make better personnel decisions to identify strengths, weaknesses, and future potential. Well-meaning efforts within the Navy and Army are attempting to combat this issues, but AI can help eliminate recurrent human problems like bias from hiring and advancement.

Companies like Adecco are already able to prescreen candidates based on skillsets, geographic preferences, experiences, and availability to open locations. Its scale is massive: AI manages their timesheets, payroll, and work prioritization for its 700,000 workers and recruiters. A similar process could be applied to initial training and follow-on unit assignments to better meet the needs of commands and services while still satisfying the lifestyle demands of individuals and families.

Retention is a problem in both the government and the private sector, but AI solutions exist which can help. This increased level of human resources personalization towards recruiting, training, and billet assignments could drastically improve morale within the services and help retain talent for the DoN. Dissatisfaction with supervisors and a lack of appreciation are two of the main reasons employees quit, and AI and sentiment analysis can help manage those effects. AI management tools also help manage workloads and burnout, which, in a military environment, could prevent catastrophes.

Administration

The routine tasks of administration with the DoN and services can be significantly augmented by existing AI services. Document preparation, completion, and handling; payment and voucher processing; policy and guidance administration; and archival storage and retrieval are all carried out by AI at varying levels within the private sector. A Harvard Business Review study found that the majority of AI projects implemented by businesses involve automating back-office tasks. These tasks include updating personnel files from e-mails and call centers, as well as extracting and updating records between multiple systems. These examples found that process-automation is the cheapest and easiest of AI technologies to implement. Administrative programs can significantly reduce the time required for manually processing high numbers of different paperwork with inaccuracies or inconsistencies, and other government agencies like NASA have already adopted these practices in some departments.

Paperwork and process automation is well-established. For instance, the consulting and accounting firm Deloitte has automated thousands of forms and saved thousands more hours of labor for its clients in the financial sector. The accounting firm KPMG partnered with IBM and Watson to learn from 10,000 documents and help its tax advisors better serve their clients. Google’s Vision OCR detects text, character, and images in documents of various file types to extract, organize, and process the relevant information. Amazon’s Textract claims to go further by creating “smart search indexes” and automated workflows for processing documents through various departments.

Today’s tech leaders – Google, Amazon, Apple, and Microsoft – each have their own AI-powered assistants which businesses can implement to streamline management and coordination. Routine work like task management, calendar management, and emails and communication can be augmented by these tools.  Businesses can also use these tools to manage facilities and systems. These same tools can be used to engage leaders and commands with their service members using 24/7 assistance. Major companies like General Electric have adopted these tools, and IBM’s Watson Assistant has led to 40 percent reductions in time spent on administrative tasks.

Transitioning these types of workloads to, or augmenting them with, AI services can reduce the time burden currently placed upon staff officers and administrative specialists. In addition to large companies like Google and Amazon, start-ups like x.ai, Voicea, and Sigrid all perform a variety of tasks like coordinating calendars and meeting schedules, setting up conference calls, managing receipts and travel processes, scheduling transportation, and scanning and saving relevant files. Communications platforms like Zoom already auto-transcribe meetings and then publish the results as text-files for easy searches.

One key way AI is changing administrative work is aiding companies in their legal and regulatory compliance. With overlapping, always changing, and sometimes contradictory sets of policies and guidance, the DoN and the services could benefit from AI tools to assist leaders at both the senior and junior levels with policy adherence. AI has allowed insurance firms, one of the most highly-regulated and complex industries in the world, to analyze documents and process claims 25 percent faster.

Savings and Possibilities

Despite the difficulties which government agencies often have when implementing new technologies, examples of effective AI adoption already exist in some areas: the review and validation of 50,000 PDF records for a federal healthcare agency, state governments achieving 100 percent compliance for land lease payments and management, and state health insurance marketplaces responding to over 1,500 customers per day. In the United Kingdom, both the Ministry of Justice and the Department for Transport have implemented AI tools to provide better services to their citizens.

Adopting AI services to aid in manpower and administrative functions will pay for themselves with an outsized return-on-investment, and free up manpower and time which the services can redirect to other specialized needs. Even a small reduction in cost can provide substantial returns. For example, a 1 percent savings in the recruiting and training budgets for the Army ($5.1 billion), Navy ($2.1 billion), and Air Force ($2.4 billion) would result in $96 million in savings. A 1 percent savings in the services’ combined administrative budgets would result in over $200 million saved. Manpower can also be reduced in these respective areas to free personnel numbers for different MOS’s or operations. For example, decreases in administrative, training, and departmental management manpower for the Army (195,500), Navy (103,800), Marine Corps (56,100), and Air Force (125,100) could allocate thousands of billets for other duties.

Conclusion

These savings and efficiency measures are even more important considering the DoN’s increased emphasis on re-allocating funding toward research and innovation. AI support for tactical military purposes certainly deserves its own attention and prioritization, but the services and their leadership must not be quick to reject the immediate benefits to be gained by AI-services in the routine and familiar worlds of manpower and administration. These implementations can provide the highest near-term benefit and make additional funds and resources available for tactical AI research or other battlefield capabilities.

Christian Heller is a graduate of the U.S. Naval Academy and the University of Oxford. He currently works as an officer in the U.S. Marine Corps, and can be followed on Twitter @hellerch.

Featured Image: Server room of BalticServers (Wikimedia Commons)

Capability Analysis

The Currency of Mission Command

1 Comment

By Capt. Bill Shafley

Information is the currency of mission command. Mission command’s six tenets include trust, shared awareness, disciplined initiative, mission orders, commander’s intent, accepting prudent risk. All trade in information to ensure an adequate combat outcome. Clausewitz urges strategic thinkers to consider the value of the object while developing ways and means commensurate with the end. Commanders and their staffs must make their information requirements explicit with Clausewitz’s reminder at the forefront of their analysis.

Subordinate units and staffs are awash in raw data. Their collective ability to add value to that data and make it actionable information is limited by capacity. Senior staffs can easily outpace their subordinates’ ability in value creation and force an investment of effort that saps capacity dry. It is a common question from senior staffs to subordinate commands to state “let us know how we can help.” There are a few ways to alleviate some of this burden.

Reshaping Commander’s Intent and Information Requirements

Subordinate commanders need clear intent. Information exchange between senior and subordinate can neither be effective nor efficient if intent is unclear or unavailable. Drafting intent is one of the most challenging of tasks for a senior commander. It involves the translation of higher headquarters’ value of the object into a theory of the fight that transforms limited resources into action. It is the commander’s opportunity to discuss risk appetite in a manner that guides effort.

Intent is foundational to decision making. It is the basis of shared awareness. If intent is left poorly defined or incomplete, neither the senior staff nor the subordinate can put to good use their disciplined initiative or begin to build shared awareness. Intent, in this sense, serves both the subordinate and the staff. The subordinate uses tactics, techniques, and procedures to convert intent into action. Staffs develop information exchange requirements based upon intent. Intent is the foundation that ties tactical action to necessary decisions. Well-conceived information exchange requirements based upon clear intent can yield success or dampen the impact of tactical action gone awry.

Commander’s Critical Information Requirements (CCIR) are a misunderstood and misapplied tool, but can be valuable to building shared awareness. A CCIR ties an information requirement, as it relates to the enemy or the friendly force, to a decision. CCIRs are developed during mission analysis. They are refined during Course of Action (COA) development and finalized through wargaming. They are eventually promulgated as part of a decision support matrix tied to events as they unfold in the tactical space.

CCIRs require action. They define what data is collected. They require some entity to collect, analyze, and report that data. A commander, or their designated representative, makes decisions based upon that information that take the outcome of a tactical action and convert it into a broader operational aim.

CCIRs are often used in a manner that mirror a set of Commanding Officer’s Standing Orders. Most naval officers are familiar with the pages full of rules of thumb, guidance, and direction to “call me when.” This type of reporting matrix has merit. It takes the guess work out of when and how to notify the commander. Yet, the “why” can get lost in the gathering of information and the notification itself. Instead of thinking tactically about the fight unfolding, leaders head down toward collecting administrative minutiae to “feed the beast.”

Planners learn very early on the “1/3-2/3” rule as it applies to time available to plan. If a tasking order is received from a higher headquarters and the confirmation brief is required in a day, then the staff has eight hours to conduct mission analysis and COA development while subordinate units have 16 hours to conduct their own planning for execution. This rule prevents higher headquarters from delving too deeply into planning for execution and leaves time for subordinate units to think through the tactics, techniques, and procedures required to execute the mission.

This same type of rule can be applied to setting CCIRs and creating the associated decision matrix. Collecting information poses a tax that trickles down into the organization, sapping the attention and effort of subordinates. At times, the amount of effort expended to gather information in response to an “ask” from higher headquarters can at best far outpace the value of the information collected. At worst, that “ask” could distract and disorient subordinates to the point where they miss a fleeting opportunity to take advantage of a developing situation. It would be best in these instances to understand that the act of gathering information takes effort and attention. Keeping CCIRs lean and necessary preserves resources of all sorts, from surveillance assets, to decision support assets, and inevitably fires. In the end, leaders may have to accept ambiguity in the short term to preserve decision space and resources in the long term.

With commander’s intent in hand and a lean set of CCIRs promulgated and understood, there is an additional tool available to trade information in a manner that supports the commander’s decision cycle. The Commander’s Battle Rhythm and its boards, centers, cells and working groups is the engine that drives operations. A disciplined approach to inputs and outputs, agendas, and membership can ensure the information being pulled into the system is fit for purpose and useable. The same 1/3-2/3 rule should apply here as well. Every opportunity should be taken to “pull” information passively into the battle rhythm and decision support matrix through official products and communications channels.

The battle rhythm should be assessed frequently to ensure the inputs and outputs are providing decision-quality information. Some of the most dangerous words uttered by a higher headquarters staff are “the boss wants to know.” These well-meaning words can impose outsized demands on subordinate units. Too many of these innocent asks could point to poor analysis on the front end. Many staffs go through the Seven-Minute Drill at the beginning of their battle staff training program. The drill is important to ensure that timings and participation are synchronized and deconflicted. Yet, the inputs and the outcomes can becomes glossed over as the battle rhythm flows through its paces. Every event is an opportunity to assess whether it is returning the proper information necessary for situational awareness and decision. A long list of due-outs and actions back to the staff or a subordinate unit may be symptomatic of the wrong inputs, outputs, and agendas.

Conclusion

Mission command’s success as a method of command and control is the commander’s business. It takes a level of commitment to and trust in the staff processes and professionalism of its members. Information requirements are the currency of command. Communication is the process in which this currency is spent. Every exchange of information is a transaction that costs something in term of level of effort. Disciplined use of CCIRs and the Battle Rhythm are manners to control the inflows and outflows of these transactions. Just like frivolous spending can create a deficit, a lack of discipline and necessity can create situations where the gains of mission command in terms of initiative can be squandered as subordinate commanders scramble to answer just one more request for information from higher that was probably right there in front of them to begin with.

Mission command won’t likely fail because of a lack of disciplined initiative or trust. It will break down because commanders have not taken the time to truly think through and generate a concise commander’s intent that codifies the information requirements necessary to build decisions upon. It will break down due to the staff’s inability to create an information bill and weigh the staff tax necessary to generate successful mission orders. It will break down because the staff just kept on asking one more innocent question, after another, and after another. It is the commanders who set the tone.

Captain Bill Shafley is a career Surface Warfare Officer and currently serves as the Commodore, Destroyer Squadron 26 and Sea Combat Commander for Eisenhower Carrier Strike Group. He has served on both coasts and overseas in Asia and Europe. He is a graduate of the Naval War College’s Advanced Strategy Program and a designated Naval Strategist. These views are presented in a personal capacity.

Featured Image: PACIFIC OCEAN (Feb. 26, 2009) Lt. Jon Bielar, left, from San Diego, and tactical action officer Lt. Paul O’Brien, from Chesapeake, Va., call general quarters from inside the combat information center during the total ship’s survivability exercise aboard the Ticonderoga-class guided-missile cruiser USS Antietam (CG 54). (U.S. Navy photo by Mass Communication Specialist 3rd Class Walter M. Wayman/Released)

Capability Analysis

The Number of Mines is Less Than Infinity

2 Comments

By Dr. Michael M. Rosenthal, Naval Surface Warfare Center, Panama City Division

“I would be very surprised if professionals engaged full time in [Mine countermeasures] MCM who speak routinely with other professionals in the same field had no genuine prior knowledge.”–Fred Huffer, Professor of Statistics, Florida State University

Mine countermeasures are actions intended to reduce the risk that mines pose to transiting vessels. Risk is defined as the probability that a transiting vessel will incur mission abort damage from a mine detonation if it travels along a predetermined route through the potentially mined area. The purpose of an MCM operation is to lower the risk so it is safer to transit. The estimation of risk is critically important to determine if the level of risk to a transiting vessel is acceptable and if the applied effort is effectively lowering the risk.

Some of the traditional MCM tactics are founded on Bayesian risk metrics that utilize a non-informative prior distribution for the number of mines. Some statisticians prefer to use non-informative priors because they feel it is more objective. The non-informative prior reduces the amount of required inputs from the tactician so there is less error introduced from the human operator, but it also defeats the entire purpose of using a Bayesian paradigm because the prior distribution does not utilize any information. It assumes that any number of mines is equally likely. In most cases, it is far more likely that there are just a few mines and it is very unlikely there are millions of mines.

Bayesian methods are ideal for the development of MCM risk metrics because they facilitate learning as information is acquired. This is good for MCM because new information and new decisions are frequently acquired throughout the operation. New events often occur which provide critical information and appropriate responses are promptly needed. These events vary greatly in detail. Bayesian analysis centers on utilizing Baye’s theorem to formally combine prior information with newly gathered information. This provides more freedom to create information synergy. 

Bayesian calculations utilize a prior distribution to incorporate information about parameters of interest before conducting a trial. In the context of MCM, the total number of mines is an unknown parameter of interest and a search or sweep of the mines is a trial. The choice in prior distribution is less critical when data is plentiful as long as the prior is not too restrictive. When there is an abundance of data, a reasonable prior will have a weaker impact on the final analysis because the posterior distribution will be primarily influenced by the data. However, data is usually limited in MCM operations, so the prior distribution will have a stronger impact on the analysis. In this case, it is critical to scrutinize the choice of prior in order for the risk metric to be meaningful.

Some of the traditional risk models actually use an improper prior for the total number of mines. An improper prior is sometimes used as an uninformative prior with the interpretation that any value is equally likely. However, many statisticians caution against using them since they are known to complicate interpretation and this usually does not provide an appropriate Bayesian update. The improper prior is formed by taking a limit of uniform distributions as the maximum number of possible mines goes to infinity. This does not converge to a probability distribution, so there is no meaningful interpretation of the prior belief for the number of mines.

Having an appropriate Bayesian update is important to MCM because if the result of a single pass of an MCM clearance does not reduce the posterior risk to an acceptable level, then the MCM staff will need to re-plan and execute a second clearance operation in the same area to further reduce the risk. This process will iterate until the risk is lowered to the threshold level.

As Fred Huffer stated, “It would be very surprising if professionals engaged full time in [Mine countermeasures] MCM who speak routinely with other professionals in the same field had no genuine prior knowledge.” If the user cannot reliably provide any information, then a large value for the expected number of mines and a corresponding large variance can be chosen because there is less certainty in that estimate. The user should be able to reliably provide some basic information that can be utilized in the prior belief by answering two simple questions:

  • How many mines could be in the region (few or many)?
    • (Few) i. e. the enemy has a small inventory or just a few can fit in the region
    • (Many) i. e. the enemy has a large inventory or they could lay thousands in the region
  • How certain are you in the amount above (low or high)?
    • (Low) a subjective guess based on limited intelligence information
      • i. e. if I were a minelayer, what would I do?
    • (High) a more objective belief based on concrete observations.
      • e. g. vessel with room for 10 mines was sighted laying mines.

The negative binomial distribution is one possible prior distribution that can be used for the mine risk Bayesian framework. Analytical expressions for the required clearance and the risk from remaining mines are easily derived. Not only is the mathematics cleaner and easier to interpret, but it also provides a sensible approach to incorporate prior intelligence information (information known before applying effort) into the analysis.

How does using the negative binomial distribution compare with the popular (uninformative) improper prior? There is no significant difference between calculations derived using the improper prior and using a negative binomial distribution with mean of 10^15 and a variance of 10^30+10^15 as the prior distribution. This default distribution implies that on average we expect there to be one quadrillion mines in the region with an unthinkably enormous variance (one nonillion + one quadrillion) to adequately accommodate our complete uncertainty in the possible number of mines prior to applying effort. To put these numbers into perspective, the surface of the ocean has an area of roughly 360 million square kilometers. That is 3.6×10^14 square meters. If you could place one quadrillion mines over a 360 million square kilometer area, the mines could be spaced on a rectangular grid roughly 0.6 meters meters apart. This setting of the prior goes as far as to say that we think it is reasonably possible that there could more than two quadrillion mines in the region.

If there is genuine concern that the end user cannot do better than this assumption, then after applying mine countermeasures a simple hypothesis test can be done to validate the user selected prior mean and variance against the default. In this way, if the end user significantly mis-specified the prior, then a flag can be automatically raised to warn the commander that the user is unable to translate the prior intelligence information into a reasonable prior belief for this operation. In this case, the command will have significant evidence to reject the assumption that a more reasonable prior belief can be derived by the end user, and return to the default condition. It is important to note that such a flag raise would not imply an outrageous deficiency in judgement by the end user. However, a flag-raise should initiate a discussion of caution toward objectively and consistently arriving at a reasonable prior distribution. Following this protocol should effectively insulate the MCM mission from this type of human error.

We generally prefer to make decisions based on an analysis that utilizes more information over a similar analysis which utilizes less information. An analysis that relies on a non-informative prior can most often be improved with an informative prior. With a non-informative prior, no prior information is being specified, so it is easy to come up with a distribution that utilizes some additional information. An informative prior often improves the interpretation and practicality of the analysis because more realistic assumptions can be made.

The updated theory in this work gently bridges more traditional doctrine into a broader realm of possibilities so that some basic information that has not been utilized can now be incorporated to improve the decision quality for the Navy.

Dr. Michael Rosenthal received his doctorate degree in Mathematical Statistics from the Florida State University in 2014. He received his bachelor’s degree in Mathematics with a minor in Statistics from the University of Florida in 2009. For five years, Dr. Rosenthal has worked at the Naval Surface Warfare Center, Panama City Division, developing basic research topics with academic colleges and assessing warfighter needs for updating and transitioning actionable tactics in the field of mine warfare.

Featured Image: BALTIC SEA (June 18, 2019) HDMS MSF-1 assigned to Standing NATO Mine Countermeasures Group One (SNMCMG1) conducts side scan sonar exercises while transiting the Baltic Sea during exercise Baltic Operations (BALTOPS) 2019. (U.S. Navy photo courtesy of NATO by CPO Brian Djurslev/Released)