PentaTorch
Supply Chain Optimization Platform

Plan and respond under uncertainty with one joint optimization

PentaTorch produces the Operational Plan for the unknown future and scenario-ready response playbooks when conditions shift, strengthening resilience while minimizing the Risk Response Cost

Planning under uncertainty—made actionable

Enterprises rarely struggle because they lack data. They struggle because the future is unknowable at the moment decisions are locked. Tariffs shift landed cost, seasonal patterns change demand and mix, transportation rates swing, and DC or supplier disruptions can instantly invalidate an otherwise “optimal” plan. The real challenge is not forecasting a single outcome. It is producing an actionable plan that remains sound when multiple levers move at once.

PentaTorch structures uncertainty into the Uncertainty Pentagon and then uses Scaled Stochastic Optimizer Engines to co-optimize decisions across the full operating system. The result is a single, integrated Operational Plan to run when the scenario is unknown, paired with scenario-aligned response plans that are ready when conditions shift, not improvised after the fact. Preparedness has a cost, and improvisation has a bigger one. PentaTorch minimizes total cost, including the Risk Response Cost, by designing the plan and the responses together. When uncertainty shifts multiple levers, the question is not “what’s the best plan in each scenario, solved in isolation,” but “what plan is best before the scenario is known.” That distinction is exactly where PentaTorch differs.

Explore why PentaTorch is different

Uncertainty is structured—not hand-waved
Volatility is modeled across the core decision levers (demand, DC capacity and availability, sell price, procurement price, transportation rates, and supplier availability) so planning reflects how the real world behaves: correlated changes, not isolated what-ifs.
Operational Plan is jointly optimized across the system
Network footprint (DC and supplier selections), inbound and outbound transportation, procurement, production, and time-phased inventory are optimized together across the Operational Plan and scenario response playbooks. Decisions remain coherent, feasible, and cost-effective under constraints.
Scenario response plans are aligned with the Operational Plan
When disruption hits or conditions shift (tariffs, demand, rates, capacity, supplier availability), response actions are already optimized to stay consistent with the Operational Plan, reducing churn, avoiding hidden feasibility breaks, and preventing risk from being pushed to another part of the network.
Minimizes the Risk Response Cost
PentaTorch minimizes the incremental cost of preparedness and response, and breaks it down coherently across inbound/outbound transportation, procurement, production, inventory, revenue, and profit.
What PentaTorch is—and what it is not
What it is
Operational Plan + Response Playbooks Optimization
One joint optimization produces: (1) the Operational Plan to execute when the future is unknown, and (2) scenario-ready response playbooks for major shifts, optimized to remain aligned with the Operational Plan.

Outputs include:
  • Operational Plan: optimized network footprint (DC and supplier selections), inbound/outbound lanes and flows, procurement, production, and inventory
  • Response playbooks: pre-optimized adjustments when conditions shift, designed to stay coherent with the Operational Plan
  • Risk Response Cost: incremental cost of preparedness and response with a clear driver breakdown

PentaTorch converts uncertainty into optimized actions across sourcing, capacity, transportation, and inventory without shifting risk elsewhere.
What it is not
Scenario-by-scenario optimization and comparison
Solving scenarios one-by-one as isolated optimizations, then comparing results, can explain exposure, but it does not produce a single plan that is optimal before the scenario is known, nor does it ensure consistency between the plan and the response actions. Solving each scenario in isolation does not produce the best plan when you don’t know which scenario will occur.
Expected-value (mean-scenario) planning
Replacing uncertainty with averages (i.e., collapsing scenarios into a single average case using means/expected values) can produce decisions that are optimal only for a fictional “mean scenario.” In practice, the plan can become infeasible or expensive when conditions deviate from the mean, because variability and recourse are not optimized as part of the decision.
Simulation
Simulation-only analysis can quantify variability, but it does not prescribe coordinated decisions. Even when it suggests a change, that change may be infeasible under constraints, break another part of the network, or simply shift cost and risk elsewhere.