Guide
Decision Workflow
Use the first solve as a baseline. From there, explore alternatives, test how the result responds to changed assumptions, and gather evidence before making a decision.
Scenario comparison
The Compare tab lets you stage Scenario B alongside your baseline. Change constraints or flow-path settings, re-solve, and see the objective delta, network diff, changed assumptions, and active-limit changes side by side.
Built-in scenario presets apply illustrative default constraint changes:
Relaxes capacity limits and tightens delivery requirements — models pushing the system harder.
Tightens budget constraints by a default percentage to model a funding cut.
Reduces capacity and slightly raises budget — models a key supplier underdelivering.
Manually adjust any constraint values for the comparison scenario.
The decision trace explains why the answer changed. If Scenario B improves the selected objective and remains feasible, you can promote it as the new comparison baseline.
Interventions
The Compare tab includes an intervention library: predefined Scenario B changes that adjust a constraint or a flow path's capacity so you can compare the impact against the baseline.
Available interventions depend on the constraint types in your model:
Raise a constrained operating capacity by 10%. Targets the top bottleneck or the first capacity constraint.
Increase available budget by 10%. Shows the modeled throughput change from the additional funding.
Reduce the minimum delivery requirement by 10%. Useful when the current target is infeasible.
Open 10% more capacity on a flow path. Tests whether a specific stage is the bottleneck.
Domain-specific templates may include additional interventions tailored to the scenario — for example, adding a purification shift, outsourcing a process step, or increasing bioreactor runs.
Select an intervention to stage it in Scenario B, then use the objective delta and network diff to assess the modeled effect of the change.
Promote a scenario to baseline
You can promote a scenario to become the new baseline. This resets the comparison point so future scenarios are measured against that configuration.
Promoting is useful after several what-if changes, when you want to continue exploring from a different comparison point.
Sensitivity summary
The Results screen splits sensitivity and exploration across Drivers, Robustness, and Explore:
Ranks constraints by leverage. When pressure-test results are ready, the ranking reflects objective loss under stress; otherwise it uses marginal value near the current solution.
Tightens one limit at a time by the selected stress level and re-solves to show whether the baseline answer stays stable, loses value, or becomes infeasible.
A 2D sweep over two constraints. Each point shows the active-limit region, with transitions and infeasible points called out so you can see where small parameter changes flip the strategy.
When two objectives compete (e.g., throughput vs. cost), the Pareto frontier shows every efficient tradeoff point. Each point on the curve represents a plan where you can't improve one objective without worsening the other.
Monte Carlo sampling over your constraint uncertainty ranges. Shows the probability that sampled scenarios stay feasible under real-world variation, and which constraints cause the most failures.
What to validate next
Sensitivity results and assumption quality together point to where validation effort pays off most. The highest-priority items are constraints that:
- -Have high pressure scores (the model is sensitive to them)
- -Have low input confidence or missing sources in input documentation
- -Are the first to cause infeasibility under stress
See Assumptions & Validation for how to fill in sources, input confidence, value type, notes, and review bounds.
Hypothesis testing
Type a plain-language "what if" and SSPLAX turns it into constraint and edge changes, previews the interpretation, and stages the reviewed changes in Scenario B.
"What if yield improves to 80%?"
→ Interprets as: increase the yield assumption from 72% to 80%. Staging it in Scenario B shows throughput jumps from 18,400 to 24,100 units/mo, with capacity becoming the new bottleneck.
You see its interpretation and any assumptions it made, so you can check them before acting on the result.
Build a decision brief
When you're ready to share findings, you can generate a decision brief that captures:
- -Baseline answer, next move, and objective value
- -Binding constraints and key bottlenecks
- -Scenario comparison results
- -Sensitivity rankings and assumption caveats
- -Narrative explanation of the decision logic
Export it or copy the narrative to share with people who don't need to open the model.