OptiU
Autonomous Optimization Models

The AI That Decides.

AOMs are the world's first closed-loop, prescriptive, adaptive optimization framework — powered by Opti, which combines Reinforcement Learning & Operations Research optimizers at the Autonomy layer. Not LLMs. Not LP/MIP. Not dashboards. 181 AOMs across five suites, orchestrated by Opti.

Where it sits

The Decision Layer is a new tier of the stack.

Above your data and your dashboards. Beside your team. Closed-loop into your transactional systems.

Layer 1
Systems of Record

Your data already tells you what happened.

Layer 2
Systems of Insight

Dashboards and BI show you what is happening.

Layer 3
Systems of Decision

AOMs decide what to do next — continuously, prescriptively, autonomously.

The Anatomy

Six components. One unit of decision.

Every AOM is composed of the same six parts. Click through the wheel to inspect each.

01Run02Decision03Explanation04KPI Impact05Alert06OverrideAOMOne unit of decision.
Component 01 · Run
auto-cycling

Cadence — when and how often the optimizer runs.

Five Core Differentiators

What makes an AOM an AOM.

The architectural commitments under every optimizer in the library.

01
Continuous Optimization

Re-optimizes in real time as data changes. Reinforcement Learning + Operations Research at the Autonomy layer — not snapshot LP/MIP.

02
Multi-Objective Trade-Offs

Profit, risk, ESG, and resilience balanced simultaneously — dynamically, not as a fixed weighted sum.

03
Closed-Loop Execution

Pushes optimal actions directly into ERP, CRM, MES, and EMS. Decision becomes execution.

04
Explainability & Trust

Every recommendation traces back to inputs, constraints, and objectives. Why & why-not — auditable.

05
Scalable Across Personas

From CFOs reallocating budgets to plant managers sequencing production, every operator gets the optimizer they need.

Not GenAI · Not OR · Not Dashboards

The category-defining matrix.

Every other AI/analytics category stops short of closed-loop optimization. This is the gap AOMs close.

CapabilityDashboards / ExcelGenAI / LLMsLP / MIP (OR)AOMs (Opti)
NatureDescriptiveConversationalStatic prescriptiveAdaptive prescriptive + autonomous
Decision logicManual intuitionLanguage-based suggestionsMathematical, staticContinuous RL policies + simulation
Handling uncertaintyNoneNoneLimited (stochastic forms)Built-in learning from uncertain environments
AdaptivityManual refreshConversational onlyMust be re-solvedSelf-learning and self-optimizing
Multi-objectiveNoneNoneWeighted-sum staticDynamic trade-offs (profit · risk · ESG)
ConstraintsUser-defined rulesWeak or implicitExplicit linear constraintsEmbedded constraint engine (CARE™)
ExplainabilityVisual chartsSummarized textNumeric outputsWhy & why-not engine
Execution loopManual actionManual promptOffline outputClosed-loop to ERP / CRM / MES / EMS
InputsHand-builtFree-form chatExpert modeling requiredConversational + auto-structured (Excel/CSV) or two-way integration
Problem Categories

Thirteen categories of decision, one optimization framework.

From the Opti Problem Encyclopedia — the decision categories Opti supports across consumers, SMBs, and enterprises.

Scheduling & Sequencing

What runs when, under setup, SLAs, and resource limits — manufacturing finite scheduling, OR block scheduling, line scheduling.

Flow & Routing Optimization

Who/what goes where across networks under cost / service / CO₂ trade-offs — last-mile, lane mix, carrier selection.

Inventory & Fulfillment Planning

Multi-echelon stock and fulfillment decisions under variability — min-max, safety stock, slotting.

Procurement & Sourcing

Multi-criteria awards across price, lead time, risk, and ESG, with MOQs and contracts honored.

Pricing & Promotions

Elasticity-aware price and promo optimization under competitive and operational constraints.

Budget / Portfolio Allocation

Selects and funds initiatives or projects to maximize benefit within skills, precedence, and budget windows.

Workforce Rostering

Fairness-aware staffing and shift design respecting skills, preferences, targeted deliverables, and labor rules.

Energy Dispatch & Blending

Unit commitment, fuel blending, and emission / cost trade-offs to meet demand at least cost within caps.

CapEx / Rollout Planning

Multi-period capital planning — telco rollout, network expansion — to maximize coverage and ROI under QoS and budget constraints.

IT / Cloud FinOps Optimization

Instance right-sizing and commitment planning across providers to hit SLOs at minimum cost.

Risk & Compliance Limits

Keeps exposure and VaR within limits via Monte-Carlo plus optimization of positions and limits.

ESG / Sustainability Trade-Offs

Optimizes cost-vs-carbon pathways; embeds emissions and policy targets directly in decisioning.

Real Estate & Facilities Optimization

Space allocation, lease and retrofit timing, and energy optimization for built assets.

By sector

Browse the actual optimizers.

Every AOM lives inside a sector value chain. Open the sector page to see what runs where.

Try one

Run an AOM in the Free Sandbox in 60 seconds.