THE 2007 AWARD WINNERS
THE AWARD FINALISTS
(Presented in alphabetical order by title, click on title to see finalist summary)
 Several of the Monte Finalists have made their presentations available online. Click on the PDF image to view the presentation. |
Best Technical Application
Best Financial Application
Best Student/Academic Project
Most Innovative Application
Most Unusual Application
Lifetime Achievement Award Winner
Dr. Narendra Soman
Master Black Belt
GE Healthcare Life Sciences
THE AWARD JUDGES
Best Technical Application:
Ray Covert, MCR, LLC
Andrew Sleeper, Successful Statistics
Karl Luce, Decisioneering, Inc.
Best Financial Application:
Huybert Groenendaal, Vose Consulting
John Charnes, University of Kansas
Jesper Johansen, Decisioneering, Inc.
Best Student/Academic Project:
Cliff Ragsdale, Virginia Tech
Thomas Grossman, University of San Francisco
Brian Malejan, Decisioneering, Inc.
Most Innovative Application:
James Wasiloff, US Department of Defense (Department of the Army)
Jay April, OptTek
Fred Ciochetto, Decisioneering, Inc.
Most Unusual Application:
Pat Leach, Decision Strategies
Chris Rae, Microsoft
Jonathon Fleck,Decisioneering, Inc.
Lifetime Achievement Award:
James Franklin, CEO, Decisioneering, Inc.
David Fredericks, Senior VP Operations, Decisioneering, Inc.
Eric Wainwright, CTO, Decisioneering, Inc.
SUMMARY OF THE FINALISTS
Best Technical Application
"Engine Removal Forecast"
GE Aviation
Jeff Heslop, Frank Gruber,
Narendra Soman,
and Alex Bogin
This project improved confidence in the estimation of the number of aircraft engine failures in a given time interval.
GE Aviation is now able to better detect an engine removal or failure rate change, so they can avoid over reacting to natural variation. This brings huge benefits to airlines all over the world and ultimately improves airlines cost of ownership in an increasingly more competitive aviation industry.
"Developing a Robust, Rugged, and Reliable Mechanical Design Using OptQuest with Minitab and Finite Element Analysis (Stochastic Multiple Response Optimization)"
Motorola
Ooi Chin Chin, Eric Maass, Wee Shou Chee, and Lim Ping Ping
Presentation
This project involved design of a two-way radio. They first used RSM (Response Surface Modeling) and FEA (Finite Element Analysis) to develop a model of the mechanical stresses at key locations on the radio. The design was then optimized using OptQuest with the model, prior to building a single physical sample. Verification was conducted that confirmed the optimal results obtained using OptQuest. OptQuest and Crystal Ball allowed them to achieve the Ruggedness, Robustness and Reliability in design, while reducing costs (by eliminating the need for least two tool modifications and sets of prototypes) and reducing the product development time accordingly.
"Hanford Soil Inventory Model"
Vivid Learning Systems
Robert A. Corbin, Brett C. Simpson, and Michael J. Anderson and Dr. Charles Kincaid ( Pacific Northwest National Laboratory)
This project was conducted for the U.S. Department of Energy (DOE), Hanford Site environmental cleanup.
This is the first time formal, comprehensive estimates of uncertainty have been generated for the inventory from liquid waste disposal sites, unplanned releases, and tank leaks at Hanford. The results provide the ability to target limited remediation resources with risk-based criteria.
"Passenger Side Inflatable Restraint (PSIR) Inflator Ballistic Performance Improvement / Gas Fill Process Improvement"
TRW Automotive
William Butler
Presentation
Customer requirements for automotive airbag inflators are dependant on multiple factors including specific vehicle configurations (size, mass etc.). To reduce Passenger Side Inflatable Restraint (PSIR) airbag inflator development time and costs, a series of experiments were performed for several inflator configurations. After using Crystal Ball and OptQuest, they were able to improved inflator gas fill parameters to meet customer ballistic specification limits while increasing production yield, improve gas fill process for more consistent fill parameters and sensitivity to inflator variation , increase gas fill process yield from 59.9% to 94.6%,and save a projected $456k.
Best Financial Application
"Cost Assessment for the Risks of a Large-scale Project: The Case of the Ariane Space Program"
Logma SA
Hervé Thiriez
Logma SA modeled all the financial consequences of the complete Ariane space program (30 satellite launchers). Prior to this work, there was no estimation of the financial consequences over 10-15 years of the complete program, and so EADS had no idea about the potential profit or loss linked to this program.
One major difficulty was that there were several hundred risks involved, which made it impossible to control without a general-purpose risk model. Logma SA developed a specialized VBA program for advanced sensitivity analysis, which resulted in monthly risk reviews, departmental cost validation, and overall cost risk reduction.
"Incremental Costs of Expanding Health Services Coverage"
Quantum Economics
Angel A. Rivera
The Union of Maritime Workers (UMW) in Puerto Rico hired Quantum Economics to estimate the incremental costs of expanding the health services of its Welfare Plan
in order to have a baseline to negotiate with insurance companies. Through Monte Carlo simulation, Quantum Economics proved that UTM can save from $1 to $2.5 million by self insurance, even if costs reach the 95 percentile.
"Quantifying Financial Risk in Business Case Proposals"
Agriculture & Agri-Food Canada
Michael Lionais
The Government of Canada has instituted a requirement that the Chief Financial Officers (CFO) of departments must personally attest to the reasonableness of financial information presented to the Government for decision-making.
If the information underlying a potential decision is flawed, executives need to factor this into their decision. Data integrity factors are used to assess the validity for the data underlying the financial information to develop a range for each cost element and Monte Carlo simulation is used to quantify the potential financial risk in a proposal. This increases the fiscal awareness of departmental executives in program development and has resulted and has caused multimillion-dollar proposals to be rewritten to reduce the potential unfunded liability to an acceptable level.
"The Surprising Influence of Bearing Type on Compressor Station Economics"
Waukesha Magnetic Bearings
Michael K. Swann
Presentation
WBC’s high-technology product, magnetic bearings, is steadily penetrating its conventional oil lubricated bearing product. To increase penetration of this higher priced product, WBC uses Monte Carlo simulation and real options analysis to show potential customers that the life cycle cost expressed as NPV and eNPV favors the magnetic bearings.
WBC’s end user customers see reduced operating expenses, and WBC sees increased overall sales as the market for its premium product increases.
Best Student/Academic Project
"Evaluation of Guarantee Structures
as Real Options using Monte Carlo Simulation"
Jicai Liu and Dr. Charles Y.J. Cheah
Nanyang Technological University, Singapore
Evaluation of guarantee structures found in infrastructure projects financed on a Build-Operate-Transfer (BOT) basis. In a BOT setting, the public and private sectors typically enter into a long-term contract (typically >10 years) that governs the design, construction and operation of an infrastructure facility.
A revenue guarantee structure found in the case of the Malaysia-Singapore Second Crossing (a toll bridge project) is valued as a form of real option using Monte Carlo simulation. A new repayment scheme is also proposed.
Proper quantification of guarantee structures will lead to more equitable risk-return sharing for both the public and private sectors.
"Financial Valuation Model of Construction Projects"
Sebastián Castañeda Arbelaez, Julio Villarreal Navarro, and Diego Echeverry Campos
Andes University
Presentation
A thesis for a Master in project management and construction engineering from Andes University. The financial evaluation of construction projects is a key tool before taking any type of investment decision; the way in which these projects are evaluated in Colombia, is not adequate from the financial and analytical perspective.
This investigation proposes a more precise way to perform a financial analysis through the use of a model that allows involving the behavior of macroeconomic and microeconomic variables that have an impact on the sector and allow know and managment the risk factors.
"Modeling Malaria Transmission with Longitudinal Data"
Derek Willis and Burton Singer
Princeton University
The most commonly used algorithm for estimating malaria transmission from longitudinal data does not work in all ecological settings. Without good estimates for malaria transmission intensity we cannot evaluate or make appropriate modifications to potential anti-malaria strategies.
The increased use of demographic surveillance systems in Africa is providing malaria policy makers with high quality longitudinal data. This simulation model will allow these policy makers to convert this data into information that will enable them to measure the impact of their current anti-malaria intervention(s) and to identify which intervention(s) are most appropriate to use in the future.
"Predicting Flight Training Devices (FTDs) at Embry-Riddle"
Kenneth P. Byrnes
Embry-Riddle Aeronautical University
Presentation
The purpose of this study was to predict the average number of Flight Training Devices (FTDs) required at Embry-Riddle Aeronautical University’s Flight Department. Simulation was preferred over optimization techniques, such as linear programming, due to the large amount of variables that affect flight training. Many of these variables are dependent on human behavior and affect scheduling, which makes it difficult to adequately predict the amount of resources required. The simulation proved that a new scheduling system should be developed and that the curriculum development team needed to develop a syllabus that balanced resource demand with teaching techniques. The simulation of the new schedule forecasted a positive $500k in revenue.
Most Innovative Application
"Capacity Modeling with Monte Carlo Simulation for Finished Goods Warehouses"
Intel Corporation
Scott J. Edwards and C. Grant Lindsay
Intel’s worldwide components finished goods warehouses adopted order processing cycle time as a key operational metric. While the metric was critical to measure operational effectiveness, it was clear that a key operations tool, the capacity forecasting model, did not enable warehouses to make the best decisions for capacity while meeting cycle time goals.
The solution was the creation of the “Next Generation Capacity Model”, a Monte Carlo simulation model with inputs defined by distributions. This new model simulates warehouse capacity (in terms of processing lines) and provides the certainty information needed to assess how many lines would most likely meet expected demand. Benefits included $15M in inventory savings,
"Domestic Distribution Network Optimization"
Maersk Logistics Inc.
Patthira Siriwan
A client in food and beverage industry requested Maersk Logistics to conduct the domestic distribution network optimization for them. Maersk Logistics needed to identify the optimal locations for distribution centers to serve all demand points at the lowest total logistics costs.
By using Crystal Ball® to run the optimization, Maersk Logistics identified USD 1.4 Million savings for the client and was able to identify the key value drivers.
"Integrated Cost / Schedule Project Risk Analysis"
Hulett & Associates, LLC
Dr. David T. Hulett
Presentation
Hulett & Associates, LLC, a project risk management consulting firm, has created an
integrated schedule and cost risk analysis to determine when a project may finish and how much it may cost.
Schedule risk is analyzed by simulation of scheduling software using other tools, and the results are brought into the cost model in Excel using CB Fit function. After the simulation, clients learn the likelihood of finishing on time and on budget and how much contingency in cost and time are needed based on the risks to schedule and cost.
Customers get a more accurate and comprehensive estimate of the risk to their schedules (some 5 – 10 years) and budgets (up into the $ billions).
"Price Optimization by using Business Risk Analysis and Game Theory"
Corvinus University of Budapest
Dr. István Fekete and Rozália Konkoly
Presentation
Magyar Telekom, a telecommunications market leader in Hungary and Macedonia and Montenegro, needed to determine the optimal price of a product, while taking into account the risks inherent in the competitive environment.
By using risk analysis and game theory together, it is possible to take into account the expected behavior of the competitors and the inter-relations among them. Then, with simulation, they can calculate the optimal price of a product and the optimal price margin, which can be used for the calculation of the list price. This information can be useful to quickly decide the range of price allowance that can be given (e.g. in case of tender negotiations). Using this methodology, companies in a highly competitive environment can retain their customers without unwanted reduction in their profit.
Most Unusual Application
"A Multiple Criteria Approach to Creating Good Teams Over Time "
Saint Joseph’s University
Ronald K. Klimberg and Kevin J. Boyle
A common problem that exists in many team based programs is the process by which the teams are chosen. This selection process is critical since teams are determined at the beginning of the class/program and may change from semester to semester or remain the same throughout the entire class/program. The authors developed a multiobjective nonlinear integer programming model using OptQuest that objectively determines team formation. This team formation model has been successfully applied for the past three semesters, for three cohorts. Responses from students and administration to the team formation process using this model, thus far, have been extremely positive, reducing stress for students and staff and eliminating significant complaints.
"Global Warming"
Individual
Gaetan ‘Guy’ Lion
Climatologists debate whether CO2 impact on global temperature is linear or logarithmic. What difference does this make in temperature increase by 2100? The author developed a simulation model that forecasts global temperature increase by 2100 using both a linear and logarithmic relationship between CO2 concentration (independent variable) and temperature (dependent variable).
The model generates probability distributions of temperature increase by 2100. The model clearly illustrates the difference it makes whether CO2 impact on temperature follows a linear or a logarithmic function.
"Statistical Analysis of Exam Scores to Determine Likelihood of Cheating on a Professional Exam: A Monte Carlo Simulation"
Bryant University
Alan Olinsky, John Quinn, and Robert DiSario
Presentation
This project was the result of a consultation in a court case involving accusations of cheating on a 100-question professional multiple choice exam with four choices for each question. As the prosecution utilized a witness who was an expert in statistical analysis, one of the authors was engaged by the defense to conduct an independent statistical analysis of the exam scores. The prosecution’s witness utilized a simulation to demonstrate, in his opinion, the relative certainty of cheating. The authors performed their own analyses, including Crystal Ball simulations to counter the testimony of the prosecution.
"World Cup Simulation"
ChungAng University
Prof. Young-Il Kim
When teaching statistics in a traditional way, we often fall in a mathematical pitfall that leads many students to a disastrous thinking that statistics is a useless and beyond their ability. To tackle this problem, we tried to build a sports model to intricate their interests. Sports models developed based on some objective data and appropriate assumptions are helpful in decision making related with sports and in expanding the sports domain. The author created a world-cup sports model using statistical simulation technique and computed all the probabilities of the events of interest in the world cup. The whole process of developing a model is shown to be linked to the statistical thinking with FIFA scores and appropriate assumptions.
Lifetime Achievement Award Winner
Dr. Narendra Soman
Master Black Belt
GE Healthcare Life Sciences
Dr. Narendra Soman has excelled in promoting Monte Carlo simulation concepts and solutions during his ten-year support of Design for Six Sigma (DFSS) development and application at GE. While employed at GE Global Research in the mid to late 90s, he worked extensively with GE’s Aircraft Engine business to develop and deploy Crystal Ball as a mainstream design engineering tool. Applications ranged from statistical tolerance analysis to robust design and reliability. GE considers Crystal Ball as one of the key tools enabling the advancement of probabilistic design.
Dr. Soman has developed and taught courses that relied heavily on the use of Crystal Ball as the primary simulation tool and developed training examples that are still in use as part of GE’s Corporate DFSS training. He has worked with a number of GE Aviation design engineers on specific applications of Crystal Ball to solve pressing problems, one of which resulted in direct cost savings over $3.5 million. He also participated in the definition of new features needed for Crystal Ball and was an alpha and beta tester of several new releases.
Recently, he moved to GE Healthcare Life Sciences and is their Master Black Belt driving DFSS implementation across this newly acquired GE business. At GE Healthcare, he has been spreading the use of Monte Carlo simulation both as a design analysis tool and as a process and product improvement tool. Narendra has many invited presentations that highlight the use of Crystal Ball in a variety of applications, notably at the 2004 Crystal Ball Users Conference, INFORMS 2004 and the East Coast Users Conference in 2006.
| 
