How Simulation-Driven Capacity Planning Actually Works

Pharmaceutical manufacturing is complex. You have dozens of resources — granulators, compression machines, blister lines, coating pans — each with different capacities, shift schedules, and changeover requirements. You have hundreds of products, each following different routes through the plant. And you have demand that fluctuates month to month, year to year.

Most companies try to manage this complexity with spreadsheets. They calculate averages, make assumptions, and hope for the best. But spreadsheets can't capture the dynamic interactions between resources. They can't model variability. And they definitely can't tell you which resource will break first when demand grows 5% per year for the next six years.

That's where simulation comes in.

The Problem with Traditional Capacity Planning

Traditional capacity planning typically works like this: take last year's production volumes, apply a growth rate, divide by available hours, and calculate utilization. If utilization is under 80%, you're fine. If it's over, you need more capacity.

This approach has three fatal flaws.

Flaw #1: Averages hide peaks.

If your average utilization is 75%, that sounds comfortable. But what if demand is seasonal? What if certain products cluster in Q4? Your average might be 75%, but your December utilization might be 110% — which means missed orders, overtime costs, and expediting chaos.

Flaw #2: Resources interact.

A compression machine doesn't operate in isolation. It needs granulated material from upstream. Its output feeds blister lines downstream. If the granulator is bottlenecked, the compression machine sits idle even if it has available capacity. Spreadsheets calculate utilization per resource, but they can't model these cascading effects.

Flaw #3: Uncertainty is ignored.

Demand forecasts are wrong. OEE varies day to day. Changeovers take longer than planned. Spreadsheets give you a single number — 82% utilization — but they can't tell you the probability of hitting 100%. They can't tell you how bad things get when Murphy's Law strikes.

Enter Simulation

Simulation models your plant as a dynamic system. Products flow through resources. Constraints propagate. Variability compounds.

Instead of calculating a single average, simulation runs hundreds or thousands of scenarios. Each scenario uses slightly different assumptions — demand 3% higher, OEE 2% lower, changeover times drawn from a probability distribution. The result isn't a single number; it's a distribution of outcomes.

This is called Monte Carlo simulation, named after the casino where probability was king. And it's the foundation of Ettala's approach.

How Ettala's Process Works

Our engagement follows four phases over six weeks.

Phase 1: Data Onboarding (Weeks 1-2)

Every simulation is only as good as its inputs. We start by collecting your operational data: resources, products, routes, demand forecasts, OEE parameters, and shift schedules.

We map your data to our schema, identify gaps, and work with your team to fill them. Missing data doesn't stop the project — we document assumptions and test their sensitivity later.

Phase 2: Baseline Simulation (Weeks 2-3)

With data loaded, we run the baseline simulation. This models your current operations with projected demand.

The output is a utilization profile for every resource — not a single number, but a distribution: mean utilization, standard deviation, 75th percentile, 95th percentile.

This reveals your bottlenecks — not just which resources are tight, but how tight, and how confident we should be in that assessment.

Phase 3: Scenario Analysis (Weeks 3-5)

The baseline tells you where you are. Scenarios tell you what to do about it.

We test operational levers (adding shifts, improving OEE, rebalancing production) and capital investments (new equipment, expanded facilities). Each scenario shows expected impact with confidence intervals.

Phase 4: Report & Recommendations (Weeks 5-6)

The final phase synthesizes everything into a report your team can act on: executive summary, baseline diagnosis, operational lever analysis, investment roadmap, and implementation guidance.

Capacity constraints don't announce themselves. They build quietly until they're crises. Simulation finds them before they find you. Get in touch to see what your plant looks like under the Monte Carlo lens.