Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 ScenGen Scenario Generation Risk Mitigation

Similar presentations

Presentation on theme: "1 ScenGen Scenario Generation Risk Mitigation"— Presentation transcript:

1 1 ScenGen Scenario Generation Risk Mitigation

2 2 What is Scengen? Imagine it like this, if two chess computers were left to play against each other for long enough they would eventually play every possible game of chess. Now what if the rules of chess could be rewritten to represent any situation like a game and then ScenGen will output all the moves you can make according to the rules. ScenGen is a unique software program that can generate all possible scenarios for a given situation at a very high speed. People will often try to think of all possibilities for a given problem but depending on the complexity of the problem can they be sure they thought of everything and can they do it fast enough?

3 3 ScenGen X-Ray (concept) Production Defect User Activity Test Coverage

4 4 Technology Applications 1) Exhaustive Testing: Mission critical computer systems such as Air Traffic Control, Nuclear Plant control, Credit Card or Trading Transaction systems have a high cost of downtime. These Systems need to be tested for all possible circumstances, ScenGen can produce that complete test set and the testing can be automated. 2) Path Optimization & Defect reproduction: When the shortest path to a particular goal needs to be found whether it’s to reproduce a defect or to locate the fastest way to get to a location via travel options. ScenGen will simply generate all possible routes and then sort them by the ones with the fewest steps first. 3) Finance Management: A Money manager can now be assessed not only against the market index or his/her peers but against the best and worst versions of themselves. ScenGen can backtrack and generate all the possible trades they could have made and compare them to the ones they did make. If the reason why they made there trades are captured then it’s possible to reduce emotional trading or bad habits based on historical experience by presenting the trader with a probability of how right or wrong they are based on past performance in making this decision for this reason. 4) White Collar Crime DNA: When a crime such as insider trading occurs and the defense is that the trader was just lucky and when they traded was a coincidence, ScenGen can generate all the other possibilities thus calculating the likelihood of this being a coincidence. An 8 million to 1 coincidence would prove to a jury beyond reasonable doubt that the trader had inside knowledge just like DNA evidence at a murder scene would give those odds that it’s someone else’s blood.

5 5 Technology Applications 5) Single Point of Failure (SPOF) Analysis: ScenGen can locate SPOFs in systems design including networks by generating all flows thru the network and then systematically simulating if each server/router of the network went offline how many flows would be impacted. 6) Disaster Recovery, Business Continuity: ScenGen can generate test and exercise scenarios for natural disaster or terrorist act combinations and ensure coverage in the emergency responders plans. 7) Cyber Security: As referred to in a) Exhaustive Testing above ScenGen can completely test a software system, Online security software from Bank Transfers to Credit Card websites are just another software system, if ScenGen can find a defect in a security system it may indicate a backdoor vulnerability. 8) Intelligence analysis: Building a model of a particular circumstance such as a business revenue model or the capabilities of the various powers in the middle east (Pakaistn, Israel, Iran, Afghanistan, etc), ScenGen can generate the possible scenarios and combinations that may play out. It's possible to weight these in terms of importance/risk too. 9) Disease, Drug or Chemicals: ScenGen can generate all possible combinations of chemical bonding, Drug conflicts or disease outbreaks.

6 6 Technology Applications 10) Financial Compliance Fraud and Abuse: Most current regulation SOX, SAS70, HIPPA is open to interpretation and wide open to circumvention as the policy is published but the implementations are all defined by the entities subject to the regulation. ScenGen can either strengthen the implementation by testing for circumventions or be applied by the Auditors to perform a more complete test. 11) Legal, Political Analysis: ScenGen can be used in white collar crime cases such as insider trading in a similar manner to DNA testing in that it can prove that the trades made and the timing of those trades is a 1 in 8 million chance of being a coincidence, thus mitigating reasonable doubt. Political analysis could benefit from ScenGen showing what the possible reactions could be for each stance on an issue. 12) Compliance: Either process and policy analysis or regulatory compliance to fully document or prove that a system has been thoroughly tested ScenGen can do the work to produce the audit evidence. 13) Impact Analysis: When changes to a project are proposed and an impact assessment as to the impact of that change on the project needs to be made then ScenGen can simply be run twice. Once on the old unchanged model, once on the new model and the results compared – the difference is the impact. 14) Education & Research: When documents and whitepapers are marked up with recommendations of which ones should be read before or after which other ones, ScenGen can generate the reading/research lists for a given subject.

7 7 Inventor’s Bio Walter O’Brien B.Sc. CSAI brings more than fifteen years of corporate, entrepreneurial & consulting experience in small, medium and enterprise business environments. Some of Walter's achievements include being ranked as the 6th fastest programmer in the world, led the national team in the Olympics in Informatics, and was granted an Extraordinary Ability EB 1-1 visa by the US government as being of National Interest to the US economy. Walter has partnered with corporate leaders to reduce risks and costs while growing revenue and efficiency at companies such as Capital Group "The American Funds", Oracle, Fujitsu, Microsoft-Trados, Baltimore Technologies and Critical Path. Walter has advanced knowledge and demonstrated success in high volume order management and transaction IT architecture, quality assurance to 99.99% uptime, process methodologies, globalization and regulatory compliance. Walter graduated with honors and holds a Bachelors of Science degree in Computer Science and Artificial Intelligence from Sussex University. Walter also holds certifications and industry training from New York Institute of Finance, JPMorgan Chase Plexus Group, CS First Boston, and HP Mercury. Walter is an active member of International Association of Software Architects and the Enterprise Architecture Forum.

8 8 Financial Management Performance AP = Actual Performance Sell S Buy B AP = 13%

9 9 Financial Management Performance BS B B B B BBBBBBB S S SSSSSSSS SSS S B BB AP = 13%BP = 18%WP = -6% AP = Actual Performance BP = Best Performance WP = Worst Performance We now know that in a range from –6 to +18 (24) The Money Manager is performing at 13+6/24 = 80% accuracy

10 10 Hopfield Neural Net No coverage here yet Dynamic weighting of the node’s business risk emulates a manual tester’s “instinct” behavior Heuristic: reduce weight of nodes as they are fired so that area’s not yet covered become relatively higher in importance Enough coverage here for now 5 5 5 4 4 3 2 1 113 7->6->5->4 5 10

11 11 ScenGen Performance & Requirements What is ScenGen’s performance? ScenGen's speed will depend on the hardware it runs on but for example on a regular 1 GHz laptop it can generate 250 human years of work in 1.5hrs, that's 1.5 million scenarios per second. What systems would it take to run ScenGen? ScenGen is very powerful but optimized to be very small (<200k) and lightweight in order to be as fast as possible. ScenGen is written in the C language and so it will run on any platform a Windows laptop, Unix, Linux, MAC, etc and has no special memory or hard drive requirements beyond space to store the scenarios generated.

12 12 Graphical Tree Output

13 13 Virtual Pointer Tree Structure RS Rules

14 14

15 15

16 16 SlipKnot Concurrency Testing

17 17 SPOF Hunting

18 18 SPOF Hunting (A,B,C to 1,2,3) A B C 1 3 2

Download ppt "1 ScenGen Scenario Generation Risk Mitigation"

Similar presentations

Ads by Google