Saturday, October 31, 2015

A Morphological Approach for Proactive Risk Management in Civil Aviation Security

By: Hernando Jimenez, Ian C. Stults, and Dimitri N. Mavris

Summary:

This article discusses a study in which morphological analysis was used in order to create a framework that would proactively assess the rest of a terrorist attack on an air transportation system (ATS). The goal of morphological analysis is to create possible attack scenarios and the likelihood of those scenarios. In this example, profiles of various terror organizations were developed and outlined so that a specific assessment of high-risk scenarios was able to be made. If this method was used by defensive organizations, then they would be capable of quickly assessing the risks of the various terrorist attack scenarios and be able to protect air transportation systems more effectively. This is a very practical example as the commercial aviation system continues to be a very critical part of infrastructure, gets a lot of attention from the media when something occurs, and continues to be a high risk target for terrorist attacks. 

This article also discusses the background of risk management and assessment. According to this article, risk is "the combination of the likelihood of a given event and the consequence or associated outcome of that event if it occurs." When the risk has been characterized and evaluated, then the next step is the mitigation of that risk. The combination of risk assessment and risk mitigation is known as risk management. Risk assessment is when the risk is modeled and quantified by estimating values for risk parameters.

When the risk was assessed in this case, it was determined as a product of probability and a consequence scalar index, then identified within a risk-level matrix that is depicted in the article. This event impact is categorized as low, medium, or high, with corresponding scalar values of 10, 50, and 100. In addition, the low, medium, and high likelihood events are modeled via probability values of 0.1, 0.5, and 1.0. For this example, the values for high risk events scored above 50, the medium risk events scored between 10 and 50, and low risk events scored below 10.

To study the vulnerabilities of the ATS in this example, the system's architecture was explored from a purely operational perspective which means that the only important elements were the ones directly involved in the architecture (i.e. control tower of the airport and the airport itself).In order to do this, morphological analysis was used. What the morphological approach allows analysts to do is to break down the potential attacks into their key components, examine and identify the combinatorial rules between them, and search for specific attacks that would reveal those vulnerabilities. The first step to do this in this specific example was to define an attack model that featured the key elements. In the model, the target is the parts of the ATS that will be primarily affected (i.e. aircraft or parts of an airport), and the tactical objective defines the impact or effect caused by the attacker. There were two limitations in the study however: the model couldn't capture ulterior motives or higher level objectives by the attackers. 

Based on the morphological field that was created by the authors, there were 5,040 different attack scenarios that were created and 1,172 of them were internally consistent. Something to keep in mind with this example, however, is that the authors of this study are not experts on ATS security, so the values and results are notional and meant just for the demonstration of this method. When the results were generated, it was found that explosives are a high-risk weapon, weapons transported in a ground vehicle or in garments or carry-on luggage were particularly interesting, and people who use airports are high-risk targets in comparison to other assets. Because of this, it seems that no single particular area of focus exists and that points of entry in areas that are more accessible and closer to passengers are at a much higher risk.

The results say that the use of morphological analysis was helpful overall. They specifically say that top-level and tactical attack models provide general applicability and show a wide range of scenarios and sensitive security information can be avoided. The fact that the results could be visually represented was also a plus since the authors say that it's vital for adequate assessment and evaluation of potential risks. It was also helpful because the attacker profiles mapped to data filters allowed the generic attack data set to be characterized and refined which then allowed a defense approach tailored to what the entities of interest were.

Critique:

Overall, I thought this article was beneficial and interesting. In intelligence analysis I can definitely see the value of using a method such as morphological analysis because scenarios are created and then different aspects of them are weighted which isn't an option with some methods. In this particular example, however, there seemed to be a few critiques that could be made which are the fact that the authors who conducted the study were not experts on ATS security which could have affected the scenarios that they were able to come up with and the fact that this method seems to be most useful for tactical analysis. It did provide a good example of a morphological analysis though, so I think it is worth the read.

Source: http://enu.kz/repository/2009/AIAA-2009-1636.pdf

Friday, October 30, 2015

MAE Design Model and Morphological Analysis


Morphological Analysis
This technique helps generate ideas. In this example, this technique helps establishes a vegetable collection system. The analysts set up a table and incorporates the vegetables and options by collating them into a single matrix. This set up helps analyst see all of the ideas together. When building the table analysts should, “try to reduce the number of generated ideas from the lateral thinking diagram into those that really are the most suitable.” The table will help the analyst visualize the various systems that will combine to the final product.


Figure 1 is the Morphological Table for a vegetable collection system. This chart shows how the Generated Alternatives are formed into a matrix with images to help supplement the variable. Analysts then begin to formulate a path in order to develop a design or scenario. This step is depicted by Figure 2.

According to the author, analysts are now at a stage to begin concept design. For starters, “the first few concept designs should follow different routes through the Morphological Matrix, that is why you have compiled it. Do not ignore it. In this instance;”
Concept 1 could be – scoop > conveyor belt > water from well > bowl > track system > wind blown
Concept 2 could be – triangular plow > rotating mover > water from well > wheel > hand push
Concept 3 could be - scoop > conveyor belt > square mesh > wheel > hand push

Analysts should be able to produce at least three well-considered concepts. These are then annotated and developed to a level that is distinguishable as an acceptable solution or key assumption.

Critique 
This method leaves a clear audit trail about how the judgments were reached and it reduces the chance the events will play out in a way that the analyst has not previously imagined. Although it, allows analysts to identify intelligence gaps, it may yield too many possibilities. Morphological analysis could also be confusing. It is effective when used early in an analytic product to generate ideas.
 

Nuclear Facilities and Sabotage: Using Morphological Analysis as a Scenario and Strategy Development Laboratory

By Tom Ritchey (Swedish National Defense Research Agency)

Summary

The study defines morphological analysis as “Morphological analysis (MA), pioneered by Fritz Zwicky in the 1930s and 40s, is a method for investigating the totality of relationships contained in multi-dimensional, non-quantifiable problem complexes”. Author argues that modelling complex socio-technical systems and developing threat scenarios comes with different methodological challenges such as the unquantifiable nature of factors, inherent irreducible uncertainties, and not enabling nature for tracing iteration of study (complex situations which is hard to reproduce the same process or results). The study suggests that MA may overcome these challenges by structuring and analyzing multi-dimensional technical, social and political problem complexes, which do not lend themselves to quantification. It can be used for developing scenarios, for defining and analyzing complex policy spaces and for assessing the relationship between ends and means in strategic planning.
MA goes through cycles of analysis and synthesis in a number of iterative steps. The iterative steps are:
Analysis phase: Define the problem complex in terms of variables and variable conditions.
Step 1: Identify the dimensions, parameters or variables, which best define the essential nature of the problem complex or scenario.
Step 2: For each variable, define a range of relevant, discrete values or conditions, which the variable can express.
The variable and variable-condition matrix is the morphological field -- an n-dimensional configuration space, which implicitly contains an outcome space for the problem complex thus defined. This outcome (or solution) space must then be defined.
Synthesis phase: Link variables and synthesize an outcome space.
Step 3: Assess the internal consistency of all pairs of variable conditions, identifying all inconsistent or contradictory pairs.
Step 4: Synthesize an internally consistent outcome space.
Step 5: Iterate the process if necessary.
In order to show these steps in application the author presents an example that assesses a nuclear plot scenario in Sweden. The author suggests that it is advantageous to develop two complementary morphological fields or laboratories: e.g. one which systematically maps out ranges of possible scenarios, based on factors which cannot be directly controlled and which put demands on the organization in question (i.e. an "external world" field); and one in which they map out alternative strategies, depending on variables which can, more or less, be controlled by the organization (i.e. an "internal world" or strategy field).
Figure 1: Segment of the “demand-field” linked to a “dirty bomb” scenario. The configuration cluster shown here is for the first time-step in the scenario: “Theft of radioactive material reported”. 



Figure 2: Scenarios linked to preparedness resource field. 


Critique

The method enables the analyst to take into consideration many variables at one time. They have already computerized it in Sweden as you see in the pictures. What the author did in this study is to lay out possible scenarios and counter-actions,and  internal demands to deal with that danger. I don't have any issue with that. However, complex phases and individual mistakes/biases while generating options or counter-measures to thriving events may compromise the success at the end. Therefore, I think, it requires some expertise to some extent to be able to conduct MA.

Source: http://www.swemorph.com/pdf/inmm-r2.pdf

Monday, October 26, 2015

Multi-Criteria Intelligence Matrices (Rating: 4.25 out of 5 stars)

Note: This post represents the synthesis of the thoughts, procedures and experiences of others as represented in the  articles read in advance (see previous posts) and the discussion among the students and instructor during the Advanced Analytic Techniques class at Mercyhurst University in October 2015 regarding MCIM as an Analytic Technique specifically. This technique was evaluated based on its overall validity, simplicity, flexibility and its ability to effectively use unstructured data.

Description:
The Multi-Criteria Intelligence Matrix (MCIM) is an analytic method that focuses on an external decision while taking into account courses of action (COA), screening criteria, and evaluation criteria. MCIM is an intelligence-focused version of Multi-Criteria Decision Making (MCDM), a process which breaks down a problem into subcomponents, evaluates each subcomponent individually, and reassembles the subcomponents in order to make a decision. MCIM provides an estimate resulting from the final scores of each COA, which are based on the evaluation criteria and represented in the matrix.

Strengths:
  • Streamlines consensus for a COA
  • When coupled with sensitivity analysis, results are strengthened (Note: this is both a strength and a weakness)
  • Doubles as an intelligence collection plan by identifying intelligence gaps and weaknesses in analysis
  • Provides a numeric score for a course of action
  • Considers multiple scenarios and their factors
  • Provides a different perspective for the analyst
  • Allows for an analyst to capture and improve upon their process

Weaknesses:
  • Qualitative data can be difficult to measure and subjective
  • There is a lot of room for mistakes to occur
  • Total weight of evaluated criteria may decrease analytical confidence
  • Is ideally suited when multiple COAs are plausible, effectiveness is limited when fewer options and evaluation criteria are available.
  • Extremely difficult to employ without subjectivity
  • Reliability of methodology is largely based on the research done for Multi-Criteria Decision Matrices

How-To:
  • Generate Courses of Action (COAs) available to your target
  • List your target’s screening criteria, what either must happen or can’t happen for the target
  • List the remaining COAs
  • Create your target’s Evaluation Criteria using numbers to weigh the COAs (i.e. Public Support--Weight: 1=Low, 2 =Moderate, 3 =High)
  • Generate the MCIM matrix (example of blank matrix found below):


Personal Application of Technique:

For the exercise the MCDM technique was utilized. We made up a simple scenario: an individual with a certain amount of budget; and this individual lives in a snowy city in the winters. This was due to the variables that would affect his/her decisions. This individual wants to buy a car and through the decision making process he/she used MCDM. Firstly, everyone in the class read through the simple scenario and then everyone listed the cars they would want to buy or any other option they could choose instead of buying a car as their Courses of Actions (COAs). Secondly, everyone listed their screening criteria. Since we conducted MCDM, the screening criteria mostly composed individual’s personal preferences. After every individual eliminated the cars/options according to their screening criteria, they formed their evaluation criteria, e.g. safety, and they assigned weightings to them as low=1, medium=2, and high=3. Finally, everyone formed their matrices and found their total points by measuring what their remaining COAs score when evaluating criteria taken into consideration. Consequently, the students found that which car/option fit to their budget, preferences etc. by looking at which COA scored higher than the remaining.