Comparing Asset Investment Optimization versus Simulation for Strategic Decisions
By Bernard Lavoie, September 2018
The difference between the terms optimization and simulation for strategic decision making is often misunderstood. There is, however, a clear distinction between the two and each of these methods should be used appropriately depending on the type of problem to be solved and the constraints involved.
As part of the implementation of a decision-making process, both approaches can be effective and efficient in establishing a tactical and strategic asset plan for areas as diverse as finance, energy, transportation or telecommunications.
As each approach typically requires the application of different tools and technology, it is important to fully understand the nature of the problem and solution in order to provide a solution that will effectively meet the expectations of decision-makers.
Optimization and Simulation Explained
Optimization is a numerical method of analysis that consists of finding the optimum, i.e. the most efficient solution in absolute terms to a problem, while respecting all the constraints involved. In asset management, it will more often be a question of maximizing or minimizing an objective function in the presence of several variables (for example, minimizing the cost to maintain a certain level of risk or minimizing risk within a cost constraint). This approach is therefore an excellent solution to sequence a set of projects over a relatively short period (for example one to three years) considering all of the constraints involved (budget, workforce, schedule, etc.). It’s always possible to run the optimization process again in order to consider the evolution of the underlying variables (e.g. delays in the work, reduction of available resources, new urgent projects, disrupted planning, etc.).
But what about the strategic aspect and the impact of short-term decisions on service levels and long-term investment requirements? In order to establish a realistic investment plan that meets the company’s short, medium and long-term objectives, it is important to consider not only the known variables, but also all of the uncertain and unpredictable variables with a significant impact on strategic decision-making in terms of funding, manpower and service levels.
Several simple examples illustrate the limitations of optimization techniques in a real asset management context:
- Which assets will require unplanned replacement due to a major break?
- Which customers will be impacted by service interruptions?
- Which assets will require corrective action next year, in two years or in five years?
An optimization approach would be very effective in considering the occurrence of such events in order to reorient the operational tactics of interventions. However, for an asset manager, establishing a strategic plan involves knowing the impact of these events and, above all, the ability to test different intervention options in order to determine the best decision-making policy that considers all known and unknown elements in the short, medium and long term.
The benefits of Simulation in Asset Management
Simulation allows for the evaluation of a large number of realistic scenarios with multiple alternative outcomes, considering billions of known and unknown variables that can impact objectives. A simulation offers considerable flexibility with a very high level of realism, better supports the strategic decision-making process, and allows the user to find the outcome with the greatest return on investment. In short, it makes it possible to determine the options that are physically feasible and financially most profitable, given the company’s expertise.
Simulation is therefore the right solution when the problem involves a high level of uncertainty, implies multiple random variables and is too complex to define with analytic expressions. The flexibility of a simulation to manage non-deterministic problems better suits the needs of organizations looking to make better asset investment decisions in the short, medium and long term.
Simulation using the DIREXYON Decision Modeling Platform
The DIREXYON platform combines combinatorial analysis, the creation of decision tree policies and the power of Monte Carlo simulation all in one software tool to support the creation of a strategic asset investment plan. By simulating a large number of realistic possible future outcomes, the DIREXYON platform avoids implied bias which can be caused by a deterministic optimization approach whereby the optimal solution is true only under predefined parameters. By digitizing and simulating enterprise expertise using reliable decision trees, the DIREXYON platform gives executives a strategic decision-making solution they can trust.