In our previous post in this series, Portfolio Review Part III, we explained the three key business benefits of cloud-based forecasting and planning, not just for Portfolio Review, but for the efficiency and effectiveness of the entire Sales and Operations Planning (S&OP) process. To bring the final curtain down on the series, we’ll conclude with this brief on prescriptive analytics.
Monte Carlo simulation
Named after the casino-gambling resort in Monaco, Monte Carlo simulation is a powerful technique for modeling the probability of different outcomes under any number of different scenarios.
Several of our large clients use our Monte Carlo simulation tools:
- Pharmaceutical companies: Predict the performance of their drug development portfolios, assessing the probabilities of sometimes hundreds of thousands of potential outcomes. They do this primarily to assess long-range revenue potential for new chemical compounds and for projects in late-stage trials. This is critical. Drug developers must continually decide where in the portfolio to invest, or how best to balance R&D and other capital investments with liquidity and other resource needs.
- Aerospace clients: Run our Monte Carlo simulations to predict costs related to multi-year support-and-maintenance contracts, and prescribe terms.
These customers set plans and policies by assessing the probabilities of myriad outcomes, such as whether internal costs will fall in a specified range, or a specific contract will lose money, or sustain a profit margin.
In most cases, clients use prescriptive analytics to assess competing courses of action and the probabilities of associated outcomes. Vanguard software can identify the sweet spot, based on management objectives and any combination of constraints. In seconds, the software can what-if test unlimited scenarios, also called scenario planning, and reveal how different decisions or circumstances will affect outcomes.
In addition to Monte Carlo simulation, predictive and prescriptive analytics (sometimes referred to as a branch of advanced analytics) include multiple forecasting techniques for very hard to predict matters, such as new product adoption, product phase-outs, and sales of products with sparse or intermittent demand. These techniques include:
- Comparable forecasting
- Spread curves
- Adoption modeling
We’ll have more on these methods in another blog post.
In the last few years, advances in sheer computing power have made exhaustively iterative Monte Carlo simulations a viable forecasting & planning tool, especially on a large, enterprise scale. To fully harness that power, we realized in our own dealings with clients that most organizations need this feature available on an automated platform, where non-technical business users on various teams can test and share plans, and coordinate actions that benefit the organization as a whole.
Over the course of two decades, we developed what is now a fully automated, cloud forecasting & planning platform that operates on a real-time mesh network. It ensures that users worldwide are always working with the latest data, so that forecasts, budgets, and plans are always in sync with any organization’s transactional database. Changes and updates are live across the entire system in seconds.
For many organizations, this is an absolute first. In fact, more than a few existing clients had been so indoctrinated in their laborious monthly forecast processes that they had difficulty understanding the full value of automated rolling forecasts, evergreen planning, and the ability to look two or three years ahead at any given time (and in a matter of seconds). More on rolling forecasts in a forthcoming post.