Risk management professionals are always looking for ways to optimize the risks in their projects and processes – that is, ways to minimize the probability of negative outcomes while boosting the probability of positive ones. In this article, we’ll explore advanced techniques for risk optimization that can elevate your project planning and decision-making strategies.
To bring these techniques to life, we turned to one of our risk management experts – Dr. José Raúl Castro, Senior Consultant and Trainer for Lumivero in the United States and Latin America. In a presentation at the Lumivero Virtual Conference, Dr. Castro presented advanced methods for optimizing risk in project management planning and outcomes using two tools in Lumivero’s DecisionTools Suite: Evolver and RISKOptimizer.
The following examples from his session show how these tools can be applied to real-world project challenges, helping teams make better, data-driven decisions in uncertain environments.
Continue reading to gain new tips and tricks or watch the webinar for the full presentation.
Applying Evolver to project management planning problems
First, Dr. Castro presented an example involving a power plant design project. The sample project had 10 tasks broken out into steps. Each step had a “prior activity” – the tasks which needed to happen before work could proceed on that step – and a minimum completion time estimate. The question Dr. Castro wanted to understand was: what was the most likely total time it would take to complete the critical path of the project?
Next, he set up the model in Evolver – a name that, he reminded attendees, comes from the words “evolutionary” and “solver” – which is part of DecisionTools Suite. Dr. Castro applied Evolver’s Minimum optimization model, setting constraints that ensured the model never returned results with lower minimum completion times than those defined by the task list.
Evolver then ran through critical path trials, respecting both the relationships between steps and the minimum completion time restraints, to return a minimum completion time of 25.
“This is a nice way of optimizing things, so that we optimize in terms of relationships,” said Dr. Castro. But this is considering one factor – time. What happens when costs are factored in, too?
Adding complexity to models for advanced optimization of risk
Dr. Castro displayed an expanded version of his project-planning model that included prices which corresponded to how long each activity took to complete. Using a basic Excel formula (horizontal lookup, or HLOOKUP), he linked costs to completion times for each task.
Next, he set up Evolver to minimize the total cost of the project while continuing to respect the minimum completion time restraints. While Evolver ran thousands of trials, Dr. Castro also demonstrated the Watcher feature, which generates a live graph showing the trend for the trials:
With the Watcher feature, you have the option to stop the calculation when it seems to have settled on a result.
The Evolver returned a minimum cost of about $4,300, demonstrating advanced optimization that considers cost factors as well as time. Project managers often need to complete their projects within a specific timeframe – a set number of days or months. Evolver can help optimize under that kind of constraint, too.
Risk optimization that accounts for timeframe reductions and soft constraints
Dr. Castro considered what would happen if a client asked to reduce the amount of time it took to complete a project by a certain number of days – say, from 45 days to 40. How would that affect costs, given that completing something faster often involves paying for overtime and other rush fees?
Dr. Castro demonstrated how to account for this additional requirement, running an Evolver optimization model that eventually arrived at a minimum cost of $11,450 to complete the project in 40 days. “We are minimizing costs, but at the same time . . . we had to spend some additional money on some of the activities to finish earlier than before,” he explained.
From here, he also highlighted how Evolver can handle soft constraints—such as flexible time or budget limits—either by allowing some leeway or by applying penalties when those limits are exceeded.
Accounting for random chance with RISKOptimizer
Cost and time aren’t always fixed, however. That’s where RISKOptimizer comes in: using Monte Carlo simulation, RISKOptimizer works on models that account for uncertainty. Dr. Castro updated his project planning model to include a certain amount of variability for the project’s costs, using @RISK’s Vary function to define the deviation range for each task.
Using RISKOptimizer, Dr. Castro was then able to run multiple Monte Carlo simulations while optimizing the mean value to arrive at the most probable cost outcome for his project if completed in 40 days: a mean total cost of $37,384. “You can see how a small model became bigger and bigger and bigger, until it eventually became a random model used with RISKOptimizer,” Dr. Castro explained.
Generate advanced optimization models with Evolver and RISKOptimizer
Interested in learning more about how you can make better-informed decisions with better risk projections? Request a demo of DecisionTools Suite today.