Supplier Integrity

Torch.AI and Deloitte Partner to Improve Supplier Integrity for Australian Army Aviation

Washington, D.C. – September 13, 2019 – Torch.AI, the leading machine learning and enterprise data orchestration platform to scale trust, announced today that it has agreed to partner with Deloitte Consulting Pty Ltd to deliver machine enabled product-trace-ability solutions to the Australian Department of Defence. The program will enhance the client’s ability to improve supplier integrity through the illumination of complex supplier networks, data discovery, ingestion, orchestration and risk analytics.  Uniquely, the platform offers continuous vetting and monitoring through the use of two Torch.AI products, NEXUS EDO and the company’s ILLUMINATION application.

Torch.AI has pioneered an evolved the approach to identifying, assessing, and neutralizing risks associated with the global and distributed nature of product and service supply chains. The global economy presents unique and complex challenges when applying risk methodologies with the goal of safeguarding government and large commercial supply chains from emerging threats and vulnerabilities. The presence and influence of adversarial foreign governments, poor manufacturing and/or development practices, counterfeit products, tampering, theft, malicious software, etc., are examples of supply chain risks that must be mitigated. Federal agencies, government contractors, suppliers, and integrators use varied and non-standardized practices, making it difficult to consistently evaluate, measure, and neutralize threats to a particular supply chain.

Torch.AI has developed several specialized software solutions that autonomously interrogate complex networks that can be concurrently persistent and ad-hoc, solving several challenges in the Supply Chain Risk Management sector.


About Deloitte Australia

The Australian partnership of Deloitte Touche Tohmatsu is committed to growth, client service and its people – 790 partners and more than 8000 people located in 14 offices across the country, plus Papua New Guinea and Timor-Leste.

To sustain its momentum, Deloitte continues to invest in innovative new services, products and people, while expanding its business through acquisitions, alliances and organic growth.

Learn more about Deloitte in Australia.


About Torch.AI

Torch.AI is on a mission to enable machine augmented trust at scale. The platform offers rapidly deployable augmented intelligence technologies that use unique data ingestion technologies and advanced techniques to provide an automation and judgement solution within a convenient framework.

Through the lens of network-centric intelligence, variable detection and change are illuminated and displayed in real-time, facilitating opportunities for improved decisioning with outputs needed to measure and optimize performance. With built-in connections for entity search, classification and investigation modules, Torch can quickly and dynamically surface out-of-range variables without manual intervention. Learn more at

Torch AI Cleared Careers

Torch.AI to Support New $75 Million U.S. Department of Defense Program to Modernize Security Clearance Vetting

Chantilly, Va.— May 28, 2019—Perspecta Inc. (NYSE: PRSP), a leading U.S. government services provider, announced today that it has been awarded an Other Transaction Agreement  (OTA) from the Defense Security Service (DSS) and Defense Information Systems Agency (DISA), to support the continued reform and modernization strategy for the National Background Investigation Service (NBIS). The two-year award represents new work for the company and has a potential ceiling value of nearly $75 million.

Under the agreement, Perspecta will work to update and rebuild the Department of Defense (DoD) personnel security vetting and adjudication technology apparatus. The goal of the program is to make the personnel security clearance process faster, scalable and more secure through an innovative delivery model. Specifically, the company plans to leverage commercial-off-the-shelf solutions combined with its advanced data analytics capabilities and expertise in emerging technology areas such as big data, cloud, artificial intelligence and machine learning, to deliver a new, more agile process to support DSS’ rapidly evolving needs.

“Perspecta is a leading DoD mission partner in supporting the development of a trusted workforce model that can meet evolving national security demands and better support the dedicated personnel involved in background investigations and adjudication,” said Mac Curtis, president and chief executive officer of Perspecta. “We applaud DoD’s efforts to drive bold, transformational change in technology and will provide an unmatched combination of proven innovation and intellectual property (IP), scalable agile development/security/operations (DevSecOps) expertise and intimate knowledge of the security clearance vetting process to our DSS customer.”

To deliver the DSS OTA program, Perspecta has partnered with Torch.AI; C3 IoT; Pegasystems, Inc.; CA Technologies (A Broadcom Company); Accenture Federal Services, LLC; Prime Technical Services, Inc.; and Next Tier Concepts, Inc.

Additional information can be found here:  Defense Security Service & Defense Information Systems Agency : Partnering with Industry to Protect National Security

About Torch.AI

Torch.AI is on a mission to enable machine augmented trust at scale. The platform offers rapidly deployable augmented intelligence technologies that use unique data ingestion technologies and advanced techniques to provide an automation and judgement solution within a convenient framework.

Through the lens of network-centric intelligence, variable detection and change are illuminated and displayed in real-time, facilitating opportunities for improved decisioning with outputs needed to measure and optimize performance. With built-in connections for entity search, classification and investigation modules, Torch can quickly and dynamically surface out-of-range variables without manual intervention. Learn more at

About Perspecta Inc.

At Perspecta (NYSE: PRSP), we question, we seek and we solve. Perspecta brings a diverse set of capabilities to our U.S. government customers in defense, intelligence, civilian, health care and state and local markets. Our 260+ issued, licensed and pending patents are more than just pieces of paper, they tell the story of our innovation. With offerings in mission services, digital transformation and enterprise operations, our team of 14,000 engineers, analysts, investigators and architects work tirelessly to not only execute the mission, but build and support the backbone that enables it. Perspecta was formed to take on big challenges. We are an engine for growth and success and we enable our customers to build a better nation.

blockchain legislation

Malta’s Blockchain Legislation: A Modern Approach to Tech Law, Ethics

On July 4th, 2018, The Republic of Malta adopted three bills establishing a robust regulatory framework for DLT (Distributed Ledger Technology), blockchain and cryptocurrency. The country, known as “blockchain” island, believes establishing clear rules and guidelines around these emerging technologies will attract entrepreneurs, innovation and technology companies. Regardless of Malta’s motives, the bills and dialogue surrounding the new legislation highlight the magnitude of the legal and ethical challenges created by technology, specifically those born out of DLT, blockchain, cryptocurrency, and artificial intelligence.

While the capabilities of robots and artificial intelligence is far from the Hollywood headlines of killer-bots wiping out humanity based on some self-generated ethics, technology is stressing our current legal frameworks and it’s time to evolve. In Rachel Wither’s Slate Article, The EU Is Trying to Decide Whether to Grant Robots Personhood, she recounts the story of a Dutch citizen whose artificial intelligent twitter-bot tweeted “at” a fashion show, “I seriously want to kill people.“ Despite the AI not liking fashion, and its response being a little freaky, it also brings up a plethora of legal questions that our current system isn’t equipped to handle. Was a crime committed? By who? Was there intent? Is anyone liable if public resources were used to respond to the threat? What if somehow the AI was able to harm people at the event, then what? I won’t go down the rabbit hole, but this real event scratches the surface of the challenges the legal industry faces resulting from disruptive technologies like blockchain and AI.



Perhaps of more immediate relevance, legal scholars now believe that autonomous artificial intelligence are close (if not already capable) of independently creating legal entities, like Limited Liability Companies (“LLC”) . Do we want legal status and rights given to organizations created by algorithms? Even if we don’t, can we stop them from having rights under current law? If AI’s are left unchecked to create, winddown, and transfer assets between entities, what stress does that place on anti-money laundering efforts, or on establishing liability in the event of a breach of contract or negligence? Even if you can establish an AI owned entity is liable, if they can quickly create new organizations in favorable jurisdictions and transfer funds electronically, how would judgements be enforced and carried out? The complexity of these challenges is not unique to AI and are shared by DLT and blockchain. The magnitude of the legal and ethical issues caused by these disruptive technologies are dizzying, leaving many scholars throwing up their arms uncertain of where to begin. However, as the issues above illuminate, we potentially risk significant disruption of our global economy if laws and ethics don’t catch up to technology, and we are long past the time for serious action.

While Malta’s legislation may lead to more questions than solutions, the effort should be commended for grappling with some of these issues and moving the conversation forward. Additionally, Malta’s legislative action creates a much-needed environment for legitimate and ethical utilizations of cutting-edge technologies to flourish and differentiate themselves from illegitimate solutions and outright fraudsters. While it remains to be seen if Malta’s efforts will stimulate significant economic activity, their efforts coupled with industry self-regulatory initiatives are critical to creating an environment where consumers and businesses can better identify those services and organizations committed to creating trust and legitimacy. Further, Malta’s leadership will force other jurisdictions across the globe to consider whether they want to participate in creating an ecosystem for legitimate modern business or operate in a grey economy. We should keep a close eye on the continued debates and discussions coming out of Malta as they just might foreshadow global debates and challenges to come.


Bayern, Shawn J., The Implications of Modern Business-Entity Law for the Regulation of Autonomous Systems (October 31, 2015). 19 Stanford Technology Law Review 93 (2015); FSU College of Law, Public Law Research Paper No. 797; FSU College of Law, Law, Business & Economics Paper No. 797. Available at SSRN:


Contributions to This Article Were Made by Clayton Pummill. Clayton is a Principal at Torch.AI focusing on legal, privacy, and cyber solutions for federal and corporate clients.  

Blockchain Technology

The Risk Blockchain Technology Poses to Supply Chain Visibility

Most nations are competing for economic stability, military superiority, and improved living standards among other reasons. Increased technological advancement has contributed to the increased rivalry between countries with China and the US being the best example of two states that are competing for superiority. In this paper, the risks posed by Blockchain technology to supply chain visibility will be addressed. As the name suggests, Blockchain technology is an economic transaction digital ledger programmed to record financial transactions and others valuable information which is linked using cryptography (Iansiti and Lakhani 04). Supply chain visibility is the availability of components, parts, or commodities in transit to be traced from the producer to the final destination with the aim of strengthening and improving the supply chain by making information readily accessible to all involved parties.

Although the blockchain technology poses numerous benefits, some of these advantages such as anonymous trust, streamlined and fast transactions may make it hard for states to track sales which possess different risks to countries. The first risk is the sale of sensitive data which may threaten the security of a nation. In most cases, attacks on a nation are mainly conspired by individuals who have support from the country’s citizens who have access to sensitive data and they sell this information on the black market which has been made easier through blockchain. Additionally, the sale of drugs and weapons has been made possible by this innovation, and it negatively affects a country’s stability.

Blockchain also encourages illegal importation and exportation between countries since people can easily purchase or sell products anonymously and effectively (Fincham 01). The first disadvantage of this move is the reduced tax for the affected states since most people who use this method evade from paying taxes which in return reduces a government’s earnings. Governments depend on taxes to finance its projects, which means that reduced taxation will hinder project implementation. For instance, if China fails to collect enough revenue to fund its mega-project, then it might be forced to seek other sources of finance which might be expensive or extend the time limit for completing the project.

Moreover, such illegal exportation through blockchain technology may harm a state’s internal market in cases where imported goods are cheaper compared to those offered internally. In such cases, most firms are forced out of business since they are unable to compete over those that have imported products to the market illegally. Additionally, those firms that use this technology to acquire cheap raw materials can produce and sell their products and services at a lower price. This move puts some firms at a disadvantage and are forced to take up measures such as shut down some of their branches, reduce wages, or lay off employees which in the long-run affect a country’s economy. In other cases, customers can use this technology to directly purchase cheap products globally a move that reduces local firms’ sales and revenues. Additionally, blockchain technology fails to specify standards to be followed during purchases which have adverse effects on a state’s economy (Palamariu).

In conclusion, it is evident that blockchain may significantly affect several logistic activities which at the long-run jeopardize a country’s economy. It is therefore essential for different stakeholders such as government, suppliers, investors, and customers among others to develop ways to regulate the use of blockchain technology with the aim of enhancing a fair and competitive market. Different states should come together to moderate the use of blockchain technology and prevent instances of illegal trade. They should implement strict international policies aimed at curbing the risks mentioned above.

Supply Chain Risk

Advancements in Battery Technologies and their Impact on Supply Chain Risk

The market for Lithium-ion batteries was 67GWh in 2016 and expected to increase in 2019 to 75GWh, which is an increase from the 5.7GWh a decade ago in the United States. The growth is due to the increased demand for the Li-Ion battery fueled by the ever-expanding areas of applications and the tremendous profits gained over the years. From the early 1990s to 2010, the battery market was dominated with the portable electronic consumers. However, the field of application expounded with the introduction of smartphones and mobile phones. More recently, the acceleration of the growth is based on the urge to revolutionize the automotive industry venturing in clean energy for their products such as powertrains and electric cars. The need for clean energy and control of cobalt overexploitation, which results in environmental degradation, has facilitated the process of power innovation. The increased demand for sustainable batteries has led to advancements in technologies aimed at efficiency while escalating the supply chain risks.

Battery Technologies

Battery technologies have resulted in the invention of super-batteries with distinctive lives and performance. The major factors that have driven the technological integration in the battery sector are the need for reliable performance and efficiency. The efficiency is the measure of battery life by the type of application. The reliability is the power it supplies in a certain time. Due to the invention of portable power consumers such as mobile phones and iPads, the need for longer lasting batteries emerge. The embedding of battery technologies in automotive sectors also enhances further research on reliable energy sources. The electric cars and the electric trains demand powerful batteries able to sustain their performance over a period. Therefore, battery technological revolution tends to optimize the life and battery power to achieve better performance (Scrosati, Jürgen and Werner p.67).

Over the years, battery technological revolution has transpired all over the world. Researchers have invested in studies aimed at improving the performance of the batteries by charging in seconds and lasting for months. Surrey University has indulged in the production of energy and storage through contact. Their main idea is to generate power through contact between two elements and harvesting. The harvested power is stored and used later. The University of California has also invented the gold nanowire batteries. The batteries are 1000 times smaller than the human hair and withstand 200,000 times recharging without indication of degradation. The cracked nanowire batteries never die. Other battery technologies include the Grabat graphene batteries, which give the electronic cars a driving range of 500 miles without a recharge. The University of Rice has invented a laser-made micro-super-capacitators technology. The battery can recharge 50 times more than the current super-capacitators and discharge slower too. These technologies are aimed at improving the performance of batteries (Pistoia p.45).

Supply Chain Risk

The increased demand for better performing batteries results into overexploitation of cobalt affecting the rates of supply. The extraction escalates the exposure to environmental hazards. Overexploitation also leads to increased air pollution resulting from wastewater drainage into the rivers. Currently, considering these factors, the government of Congo has decided to shut down the Katanga Cobalt mines leading to low supply (Eichstaed p.43). Shutting down the mines renders people jobless leading to low living standards with raised cost of living. According to (Eichstaed p.56) the cost of living in Congo is 147% higher than in the United States. Since machines need batteries to operate, the government may have divided opinion on reinstating such mining activities considering their adverse effects on the environment. Therefore, the major Cobalt supply chain risk is overexploitation in contrary to the mining policies especially in Congo (Huggins p.23).

In conclusion, the demand for lithium-ion batteries has increased the in the United States in the past decade. Battery technologies aim at producing super-batteries with high-performance power and prolonged lives. The battery industry has undergone numerous technological revolution leading to the introduction of batteries such as the Grabat graphene batteries, which provides 500 miles drive for electronic cars without a recharge. The significant Cobalt supply chain risk is the overexploitation in contrary with the government policies, such as in Congo, leading to environmental degradation.


Scrosati, Bruno, Jürgen Garche, and Werner Tillmetz. Advances in Battery Technologies for Electric Vehicles. , 2015. Internet resource.

Pistoia, G. Lithium-ion Batteries: Advances and Applications. , 2014. Internet resource.

Huggins, Robert A. Advanced Batteries: Materials Science Aspects. New York: Springer, 2015. Internet resource.

Eichstaedt, Peter. Consuming the Congo: War and Conflict Minerals in the World’s Deadliest Place. Chicago: Lawrence Hill Books, 2018. Internet resource.

Network Economics

Network Economics in the Era of Artificial Intelligence

In its primary context, a network is a foundation upon which humans are interconnected to each other in what they do. In the globalized world, the primary issue is the consideration of the numerous choices that people and businesses have to undertake in the information era. The origin of the understanding of network economics is traced back to the classical work of Cournot (1838). The theorist was the first economist to explicitly state the relationship between the competitive price where there is an intersection of the demand and the supply curves. Another scholar who postulated the idea was Pigou (1920), who described it in the perspective of setting out a transportation network that comprised a system-optimized and a user-optimized solution.

In the present day, the emergence of artificial intelligence means that humans have awakened to the reality of machine learning where information is now perceived in a more computerized manner. One key area that has been of focus in the concept of the “strength of weak ties” was postulated by sociologist Mark Granovetter in 1973. It is the primary basis in the analysis of social networks, especially in the process of linking the micro and macro entities in the sociological theory (Granovetter 1360). A more challenging theme that has come up with the emergence of network economics in AI is connected by six degrees of separation. The model postulates that anyone on the planet can link up to another person in six steps. It follows that when one is connected in a given dimension, there is a chance that there is more linkage than they can perceive.

The process of networking requires that the elements of strong and weak ties are both factored because even though they perform varying functions, they extend the potential beyond the reasonable reach. When formulating the theory, Mark Granovetter describes that there are various interpersonal theories that exist between disparate groups and that these ties constitute what holds different units of the society. Humans thus tend to multiplex relationships so that they represent weak ties to some of their connections and strong ties when they link with others. It is, therefore, comparable to a network multiplexer that has varying relations and that constitutes diverse types of signals.

According to the theorist, the relevance of these ties is perceived in social networking.  A strong link is thus viewed in economics as a group of geeks who are conversant with what is expected of them in a given field, such as clinical or science. They are always abreast of the information as it comes and is informed of what information is happening and going in the given field that they have specialized in. The subject of weak ties thus results from the apprehension that it is tenuous forms of relationship where they do not seem to be much conversant in clinical or on the particular scientific field that is being discussed. Despite being on the edges of influence, they are not informed of the advances in health and clinical science issues. It is worth noting that according to the theorist, these characters are crucial because they form the building of strong ties group together through the effect of bringing circles of contact in a central place and the process strengthening the existing relationships. They are important because as a result of their presence, it is possible to share the information on clinical issues and scientific trends between the different groups.

The other application of the theme of network economics is in the concepts of Artificial Intelligence (AI). The pertinent example is the case of Artificial Neural Networks (ANNs) where it is described as the powerful relation through the use of multivariate tools for dependence analysis. They have initially been applied in the neuroscience, but have recently gained media attention especially, in economics and finance. The significance of ANN therefore is that it can be used for modeling purposes and in the prediction of outcomes, because it uses machine language. These associations are relevant, because while the goal has always been to improve and replace the use of manual processes through automation, much had not been explored on the possibility of designing machines that demonstrate intelligence comparable to that of humans’. The realization that humans could as well multiply their human intelligence through artificial means has thus ended in major advancements in the civilized world. The concept of strength and weak ties is especially important in ensuring that bindings relationships are established that are long-lasting as it is the principle of this form of network. Thus, artificial intelligence has been of great advantage, because it has been possible to bring out the emotional quotient into machines with many appreciating the advancements. With the far-reaching applications that have been witnessed, it is intriguing to think of how much impact there will be in the coming decades and years.


Granovetter, Mark S. “The Strength of ‘Weak’ Ties.” American Journal of Sociology, 78, no. 6, 1973, pp. 1360–1380.

Applied Behavioral Modeling

Machine Learning in Applied Behavioral Modeling


After researchers had done specific studies to understand the human action and patterns of life including their environments, lives, behaviors, and motivations, they need to know how to present the obtained information and create a design that will result in successful representation. Individuals act differently from one another; thus, it is possible to collect different conversations and observations from various people to access human behavior and action. The user profiling and modeling are some of the examples that have been used as an evaluating system to predict the user’s behaviors for a given period. In user profiling, the personas represent various types and groups of the subject to enable the designers to develop appropriate solutions in reiterative processes. Although clear guidelines for using computer derive ad-hoc dynamic persona-types to classify life patterns and behaviors has not been established, an idea can be developed to guide on the same. The paper aims to reflect on the concept of using computer derived ad-hoc dynamic persona-types that relate to social networking, experience, and human actions.

Applied Behavioral Modeling

The construction of the profile and applied behavioral modeling for the users is based on studying their behavior patterns, cognitive features, and demographic data. Such features help provide a practical approach to represent the user’s interests and preferences. The focus of such an ad hoc dynamic involves assessing the interactions of the user with a system and do not deal with complex social networking like educational hypermedia or focus in serious games. Most of the computer-derived models are created to describe market behaviors, and they use personas or user models for representation. The model users provide a precise way to think and communicate about how the persona think, behave, what they wish to achieve and why (Fernandez-Llatas et al. 15434). The motivations and behaviors of the persona are observed and represented throughout the design process of a computer-derived model. The persona must be regarded with a considerable sophistication because using their stereotypes or generalization would not be enough to produce a clear representation. Besides, discretion and vigor have to be applied to identify the meaningful and significant patterns in the user’s behaviors and utilize the acquired information represent a broad cross-section of the persona. The dynamic information modeling focuses on personality and diverse computing experiences.

Further, utilizing the dynamic information of the user could be used to create a system that can adapt to the user dynamically. The ability of the system to adapt to the user is essential in identifying and highlighting potential users as well as predicting their behaviors. Thus, it is significant to understand that dynamics in modeling deals with lifestyles, ages, IT consumption, and space. The dynamic information is substantial, especially in the studies involving ceremonies interactions and social activities involving teenagers (Fernandez-Llatas et al. 15436). Ceremonies, especially the traditional ones provide typical social activities that enhance the sense of belonging to the members of the family, which are passed through generations. Thus, the custom is inheritable, meaning it will be predictable in the future as the young people grow up to become the target audience of such traditional ceremonies. Therefore, if the designers of the computer-derived ad-hoc dynamic can identify such potential from the users and derive data from the right source, they can support some observations like cultural heritage.

Another suitable example used in computer-derived models is the social interactions among teenagers, mainly through technology usage. Studies have established that although the youths are actively using technology to look for new friends on social media platforms, the economic status can limit their modes of communication. Generally, teenagers have limited finances, and most of the IT products or media platforms offer such services at a certain fee (Fernandez-Llatas et al. 15434). However, the older teenagers can find financial freedom later in life and develop heightened attentiveness and dependence on the Internet; thus, making them the primary target audience of the IT products and consumption, which is a concept that can be applied to predict the near future.

Benefits of technologies such as graph database in sociology

Technological advances have developed beneficial programs that do not require static patterns to process event data, particularly that of human behaviors and actions. For instance, the graph database has been used for data storage and representation. The key concept of this database is the graph, edge or a relationship of the observations and behaviors that relate the data items in storage directly. The relationship represented in the graph database allows the stored data to be linked or combined it to create a successful representation (Huang et al. 3). Besides, process mining technology enable the sociologists to facilitate workflow interpretation from certain event records and reports while conducting studies. This technology interprets graphs that are understandable by the experts studying human behavior patterns using the routine actions recorder by ambient intelligence environments. Thus, it is easier for the experts to comprehend the process of human action as well as deduce a comparison using the previous inferences to identify particular behavior patterns or changes.


The process of analyzing human behavior patterns is extensively used for several research fields, especially in sociology. Most sociologists consider the use of IT and age as dynamic attributes of the user’s profile while conducting a study to classify behaviors and pattern of life. The applied user’ profiles and applied behavioral modeling should reflect their changes in hypermedia experience as well as behavioral changes based on demographic settings and interests. Thus, the research design using computer derived ad-hoc dynamic persona-types to classify behaviors and pattern of life has to consider the anticipated changes in lifestyle, age, IT consumption, and economic status of the user.


Fernández-Llatas, Carlos, et al. “Process mining for individualized behavior modeling using wireless tracking in nursing homes.” Sensors 13.11 (2013): 15434-15451.

Huang, Ko-Hsun, Yi-Shin Deng, and Ming-Chuen Chuang. “Static and dynamic user portraits.” Advances in Human-Computer Interaction 2012 (2012): 2.

Augmented Intelligence

Torch.AI Recognized By GovCIO Outlook for Augmented Intelligence

Torch.AI, a leading provider of intelligent systems of trust for the federal government and the fortune 500 is proud to announce being featured in Industry Magazine, the GovCio Outlook.

Torch.AI was featured by the magazine following the company’s ability to offer a machine learning platform that enables complex systems to be managed and manipulated with unparalleled security and ease of use. Torch’s computing technology leverages asymmetric encryption in containing and computing data securely, in a format that is can be shared, authenticated, credentialed, and security is handled in a simple package.

The ability of the company’s platforms to store data tags securely, rather than data itself, eliminates the need to have a large data warehouse as well as computation limitations. Users can also customize the platform to ensure key variables to surface quickly. According to Brian Weaver, Founder and CEO, “The most rewarding part is making heroes out of our clients when processes are triggered based on the specific patterns of data detected by the platform. Taking their expertise and putting it into action.”

The unmatched expertise of the company in the deployment of OEM applications has attracted several consultancy firms such as Deloitte, Accenture, DXC and many others. The partnership between them has been a vital aspect that has led to the development of state-of-the-art solutions for well-known organizations such as USDA, DISA, Treasury, FSB, FNS, IRS and DoD.

Torch is currently leveraging on staff in Beltway and R&D to add value to its client’s businesses by reducing improper payments and eliminating insider threats. The company also pursues meaningful relationships with key state contractors as pointed out by Mr. Weaver: “Our partners are thought-leaders and have true domain expertise that makes an advanced platform like Torch, really shine.”

About Torch.AI

Torch.AI is on a mission to enable machine augmented trust at scale. The platform offers rapidly deployable augmented intelligence technologies that use unique data ingestion technologies and advanced techniques to provide an automation and judgement solution within a convenient framework.

Through the lens of network-centric intelligence, variable detection and change are illuminated and displayed in real-time, facilitating opportunities for improved decisioning with outputs needed to measure and optimize performance. With built-in connections for entity search, classification and investigation modules, Torch can quickly and dynamically surface out-of-range variables without manual intervention. Learn more at

About GovCio Outlook

GovCio Outlook is a technology magazine whose focus is the trends, opportunities, and challenges for CIOs in an effort to deliver efficient technology-driven services and operations to enable smart governance.

Machine Learning

Machine Learning and Artificial Intelligence

A recent article on Torch.AI as featured in Entreprenuer Magazine, provides insight on how a straightforward approach to machine augmented decision making can boost a company’s revenues and protect them from fraud. According to the article, advancement in technology has made correlation and data the most precious resources and has changed the basis of competition in addition to the creation of new paths for companies across all industries (“How a Straightforward Approach to Machine Augmented Decisioning Will Protect Companies From Fraud and Boost Revenue,” 2018). Companies are losing potential revenue from advances in technology due to lack of skills to extract sustainable and meaningful value from small sets of data. Also, companies lack the needed infrastructure to incorporate, organize, and collect data insights. Therefore, companies such as Torch.AI has developed enterprise augmented intelligence solutions to assist firms that struggle with data management (“Home – Torch.AI: Home,” 2018). Adoption of technologies, such as machine and blockchain technology, innovative and new avenues are explored to solve the issues.



Torch.AI has created a collaborative team of business analysts, solution architects, and advanced technologists to develop advanced software applications that make data more relevant, intelligent, and usable to people and provide solutions to problems addressed in the article (“Home – Torch.AI: Home,” 2018). Machine learning and artificial intelligence are being used in the current world to power simple things. Gmail, for example, uses artificial intelligence to filter most of the spam emails. Artificial intelligence can convert unstructured data into network-centric structured information that can be used to solve complex problems faced by firms in various industries. According to the SV Advisory Group, hard issues facing clients can be addressed using complex technologies presented in a simplified form. Notably, strategic use of technology has the chance of boosting a company’s revenue (“How a Straightforward Approach to Machine Augmented Decisioning Will Protect Companies From Fraud and Boost Revenue,” 2018). The high cost of labor is the primary challenge for many firms, but through artificial intelligence and machine learning, cost-effective physical robots are developed, and they increase productivity, improve accuracy and predictions, and expand discovery of solutions to complex problems. Machine learning can mitigate fraud by using network-centric intelligence to eliminate insider threats and reduce payment errors. Torch.AI has partnered with technology consulting firms and commercial enterprise customers to enhance service delivery through the incorporation of artificial intelligence and machine learning (“How a Straightforward Approach to Machine Augmented Decisioning Will Protect Companies From Fraud and Boost Revenue,” 2018). The company has adopted AI implementation by appreciating its ability and using it to make appropriate use of information through various machine learning applications and complex neutral network analysis.

Defining Risk

Defining Risk

In ordinary conversations, a risk is the likelihood or possibility of something usually negative happening. For example, the risk of getting involved in an accident while driving. But in the context of engineering, a risk is used to express both the likelihood of occurrence, for instance, the failure of a structure and the degree of consequences that it may result to, for example, loss of lives. The connection of probability and consequences is used to assess relative risks through quality and judgment which is expressed by high, medium and low. In the case where the likelihood of occurrence and the consequence is quantified, then the product is the risk(Modarres, 34). Therefore risk equals the likelihood of occurrence times the consequence of the occurrence.


Risk quantification is a process where we assess the risks identified in order to come up with data that can be used when determining a response to conforming risk. According to PMBOK standards, risk quantification is the second step after carrying out risk identification and before developing the response and control of a project risk management. PMBOK defines risk quantification as a way of assessing risks and their interactions to evaluate the extent of possible outcomes(Hillson, 33). Generally, quantification of risk is a process of assessing the risks already identified and coming up with data that will be required in deciding what is needed to be done about them.


risk quantification


The aim of carrying out risk quantification in a project is to formulate possibilities in terms of time, human resources or costs and prioritize them according to their likelihood and severity in order to come up with proper actions accordingly. It also assists companies in coming up with appropriate decisions in case of uncertainty. Risk quantification also provides assurance when dealing with unanticipated events in future instead of behaving unreasonably. When quantifying risk, we should identify them first, then analyze them according to their likelihood of occurrence and how it will impact the outcome. The likelihood is based either on perception or data of previous failure rates accessible for comparable events in a datasheet(Hillson, 45). After calculating the likelihood of all the events, we define the criterion of the probability of the entire events.

If a particular event occurs in outstanding situations, such that it is less than 3 percentage the chances of its occurrence, then its probability can be identified as rare. Likewise, consequences or relentlessness of the events on a project is also categorized. For example in a situation where the event might lead to rejecting the whole project then the project can be categorized as Catastrophic. If it results to an additional or a negative of 50 percent of the initial schedule cost then it can be categorized as major. Therefore risk is calculated by multiplying the impact or intensity by the likelihood of occurrence.

After quantifying risk, they are then evaluated in a risk matrix where the red zone represents unacceptable risks, while acceptable risk is represented by a yellow zone and neglectable risks are indicated by a green zone. For instance, if the probability of an event is classified as likely and its severity classified as Catastrophic, in this case, it will be grouped at the red zone in a risk matrix(Modarres, 51). This means that the risk is unacceptable and requires immediate attention to reduce the risk into the acceptable zone or formulate possibilities.

The cross-entropy (CE) methodology is a Monte Carlo way of sampling and optimization. The method is used to estimate probabilities. Monte Carlo is a computerized mathematical simulation method that is used to quantify risk in project management. The technique can be used to score or identify risk in a global trade or a supply chain. It is also helpful when analysing the likely outcomes of decisions and evaluating the intensity of risk that can be used in decision making. In each event, the most probable and the least probable approximations of risk are provided and then added to compute a range of possible outcomes. After that Monte Carlo analysis then produces random figures among the range and computes how many times the figure lies in each probable outcome (Concept of Risk Quantification and Methods Used in Project Management – Apppm). This likelihood is then distributed and a conclusion is made according to the most likely outcome.


risk monte carlo simulation


For instance, if a project requires three tasks, the best case and worst case approximation of the task as shown in fig 1 below. The figure shows that the project is probably going to be completed within 11 and 23 days. For instance, if Monte Carlo simulation is done 500 times producing random figures within 11 and 23, then the sum number of periods the stimulation results was less than or equal to the projected duration is calculated. After that, the probability of each projected duration is calculated and distributed as shown in figure 2. In reference to figure 1, the most probable time to complete the project is 17 days while according to figure 8 the probability of completing the project within 17 days is at 33percent although the probability of completing in 19 days is 88 percent. Therefore, it can be approximated that it will take 19 to 20 days to complete the project.

To sum up, when defining risk we can use metaphors to communicate new or broad concepts, for example, a risk is like brakes in a car, your brakes are supposed to work perfectly in order for you to reach your destination safely. No one will board a car if they were informed that the brakes do not function properly. We should also see risk as cockroaches, they hide, multiple fast and are a problem to stomp out.


Concept of Risk Quantification and Methods Used in Project Management – Apppm.  Accessed 22 Oct. 2018.

Hillson, David. Managing Risk in Projects. Ashgate Pub., 2009.

Modarres, Mohammad. Risk Analysis in Engineering. Chapman and Hall/CRC, 2016.