How Demand Forecasting Works
Understand how Prognosis forecasts patient demand, plans production, and identifies supply risks for your clinical trial.
Overview
When you complete the trial wizard, Prognosis analyzes all your inputs and generates a comprehensive supply chain forecast. This guide explains what happens behind the scenes and how your inputs affect the results—helping you get the most accurate forecasts possible.
The Three Pillars of Forecasting
Prognosis forecasting rests on three interconnected analyses:
1. Patient Demand
What it answers: "How many kits do we need, and when?"
The system projects when patients will enroll, when they'll have visits, and what kits they'll need at each visit. This creates a timeline of kit demand across your entire trial.
2. Production Planning
What it answers: "When should we manufacture, and how much?"
Based on the demand timeline, Prognosis works backwards—accounting for shipping times, manufacturing lead times, and shelf life—to recommend when production should occur.
3. Risk Assessment
What it answers: "Where might we run into problems?"
The system compares projected demand against available supply to identify potential shortfalls, helping you address issues before they impact your trial.
Understanding Patient Demand
How Enrollment Projections Work
When you enter your recruitment dates and patient targets, Prognosis distributes those patients across your enrollment window. The pattern depends on the recruitment curve you select:
| Curve Type | Best For | What It Looks Like |
|---|---|---|
| Linear | Steady, predictable enrollment | Even distribution throughout the period |
| Bell Curve | Typical trial enrollment | Slow start, peak in the middle, gradual finish |
| Fast Bell Curve | Quick ramp-up trials | Rapid early enrollment, longer tail |
| Slow Bell Curve | Cautious enrollment | Gradual ramp-up, faster finish |
| Custom Monthly | When you have specific targets | Follows your month-by-month plan |
Pro Tip: If you're unsure, the standard bell curve reflects how most trials actually enroll—slow at first as sites activate, peaking as the trial matures, then tapering as targets are reached.
How Visit Schedules Drive Demand
Once patients are "enrolled" in the projection, Prognosis schedules their visits based on your treatment arm configuration:
- Each patient follows the visit schedule you defined
- At each visit, they receive the kits you specified
- Drop-out rates reduce the patient count over time
This creates a detailed picture of exactly when kits are needed, by whom, and where.
For In-Progress Trials
If you've entered actual enrollment data in the Actuals step, something powerful happens:
- Confirmed patients are locked in — Real data replaces projections for past dates
- Remaining patients are redistributed — The forecast adjusts to hit your targets with remaining time
- You get a hybrid view — Actual history plus intelligent projection
This means your forecast gets more accurate as your trial progresses.
Understanding Production Planning
Working Backwards from Demand
Production planning starts with a simple question: "When do patients need kits?"
From there, the system works backwards:
Patient needs kit on June 15
↑
Kit must arrive at site by June 10 (buffer time)
↑
Kit must ship from depot by June 1 (shipping lead time)
↑
Kit must be at depot by May 25 (processing time)
↑
Production must complete by May 15 (depot transit time)
↑
Production must START by April 15 (manufacturing lead time)
This backward calculation ensures kits are ready when patients need them.
What Affects Production Timing
Several factors influence when production runs are recommended:
| Factor | How It Affects Timing |
|---|---|
| Manufacturing lead time | Longer lead times mean earlier production starts |
| Shipping lead times | Distant countries need earlier shipments |
| Shelf life | Short shelf life may require more frequent, smaller batches |
| Depot network | Complex networks add transit time |
Accounting for Your Existing Supply
If you've entered:
- Custom lot configurations — Pre-planned production runs
- Existing inventory — Kits already in your supply chain
Prognosis factors these in first, then recommends additional production only for the gap.
Site Seeding
The system also plans for initial site stock:
- Sites need kits on hand before the first patient arrives
- Seed stock may need refreshing if it approaches expiry
- Seeding quantities are based on your configuration
Understanding Risk Assessment
What Creates Supply Risk?
A "risk" occurs when demand might exceed available supply. Prognosis identifies several types:
Timing Risks
What it means: Kits won't arrive in time to meet demand.
Common causes:
- Long shipping routes combined with tight enrollment windows
- Unexpected enrollment acceleration
- Manufacturing delays
How to address: Adjust production timing, use faster shipping, or build in more buffer.
Depletion Risks
What it means: Earlier demand used up supply that later demand needs.
Common causes:
- Higher-than-expected enrollment in early months
- Insufficient production quantities
- Inventory consumed faster than projected
How to address: Increase production quantities or add additional runs.
Expiry Risks
What it means: Kits will expire before they can be used.
Common causes:
- Short shelf life products
- Uneven demand (large early production, slow late enrollment)
- Inventory sitting too long in the supply chain
How to address: Smaller, more frequent production runs; better demand alignment.
Reading Risk Results
When Prognosis identifies a risk, it tells you:
- What: Which kit is at risk
- Where: Which country or depot
- When: Which month the shortfall may occur
- How much: The size of the potential gap
This gives you actionable information to make adjustments.
Getting Better Forecasts
Input Quality Matters
Your forecast is only as good as your inputs. Here's what has the biggest impact:
| Input | Impact on Forecast |
|---|---|
| Patient numbers | Directly scales all demand |
| Enrollment dates | Determines when demand occurs |
| Visit schedules | Determines frequency of kit needs |
| Lead times | Affects production timing |
| Shelf life | Affects batch sizing and timing |
Common Pitfalls to Avoid
Overly optimistic enrollment: If you assume faster enrollment than reality, production may start too early, risking expiry.
Forgotten lead times: Underestimating how long shipping really takes can cause timing risks.
Ignoring shelf life: Long production runs of short-dated product create waste and risk.
Static assumptions: Trials evolve—update your forecast as plans change.
Tips for Accuracy
- Use historical data — Base enrollment curves on similar past trials
- Validate lead times — Confirm shipping estimates with your logistics team
- Build in contingency — Overage percentages protect against variability
- Update regularly — Refresh forecasts as you learn more about actual performance
Frequently Asked Questions
How often should I update my forecast?
For planning trials, update when assumptions change significantly. For in-progress trials, enter actuals weekly or bi-weekly to keep projections current.
Why does my production recommendation look different after I entered actuals?
With real enrollment data, the system recalculates remaining demand. If actual enrollment differs from projections, production recommendations adjust accordingly.
Can the system handle enrollment faster than projected?
Yes, but you may see timing risks appear. The system will flag months where accelerated demand might outpace supply, giving you time to react.
What if I have complex depot networks?
Prognosis handles multi-tier networks (central → local → country → site). Each leg's lead time is factored into the total time from production to patient.
How does shelf life affect my forecast?
Short shelf life products require careful timing. The system ensures production arrives early enough to ship and be used, but not so early that expiry becomes a risk.
Summary
Prognosis forecasting helps you answer the fundamental supply chain questions:
- Demand: How many kits, when, and where?
- Production: When to manufacture, and how much?
- Risk: Where might problems occur?
By understanding how your inputs affect these outputs, you can create more accurate forecasts and make better supply chain decisions for your clinical trial.
Next Steps
- Demand Analytics — Learn how to read and use your forecast results
- Scenario Planning — Test different assumptions to optimize your strategy