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.
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.
Your data already tells you what happened.
Dashboards and BI show you what is happening.
AOMs decide what to do next — continuously, prescriptively, autonomously.
Six components. One unit of decision.
Every AOM is composed of the same six parts. Click through the wheel to inspect each.
Cadence — when and how often the optimizer runs.
What makes an AOM an AOM.
The architectural commitments under every optimizer in the library.
Re-optimizes in real time as data changes. Reinforcement Learning + Operations Research at the Autonomy layer — not snapshot LP/MIP.
Profit, risk, ESG, and resilience balanced simultaneously — dynamically, not as a fixed weighted sum.
Pushes optimal actions directly into ERP, CRM, MES, and EMS. Decision becomes execution.
Every recommendation traces back to inputs, constraints, and objectives. Why & why-not — auditable.
From CFOs reallocating budgets to plant managers sequencing production, every operator gets the optimizer they need.
The category-defining matrix.
Every other AI/analytics category stops short of closed-loop optimization. This is the gap AOMs close.
| Capability | Dashboards / Excel | GenAI / LLMs | LP / MIP (OR) | AOMs (Opti) |
|---|---|---|---|---|
| Nature | Descriptive | Conversational | Static prescriptive | Adaptive prescriptive + autonomous |
| Decision logic | Manual intuition | Language-based suggestions | Mathematical, static | Continuous RL policies + simulation |
| Handling uncertainty | None | None | Limited (stochastic forms) | Built-in learning from uncertain environments |
| Adaptivity | Manual refresh | Conversational only | Must be re-solved | Self-learning and self-optimizing |
| Multi-objective | None | None | Weighted-sum static | Dynamic trade-offs (profit · risk · ESG) |
| Constraints | User-defined rules | Weak or implicit | Explicit linear constraints | Embedded constraint engine (CARE™) |
| Explainability | Visual charts | Summarized text | Numeric outputs | Why & why-not engine |
| Execution loop | Manual action | Manual prompt | Offline output | Closed-loop to ERP / CRM / MES / EMS |
| Inputs | Hand-built | Free-form chat | Expert modeling required | Conversational + auto-structured (Excel/CSV) or two-way integration |
Thirteen categories of decision, one optimization framework.
From the Opti Problem Encyclopedia — the decision categories Opti supports across consumers, SMBs, and enterprises.
What runs when, under setup, SLAs, and resource limits — manufacturing finite scheduling, OR block scheduling, line scheduling.
Who/what goes where across networks under cost / service / CO₂ trade-offs — last-mile, lane mix, carrier selection.
Multi-echelon stock and fulfillment decisions under variability — min-max, safety stock, slotting.
Multi-criteria awards across price, lead time, risk, and ESG, with MOQs and contracts honored.
Elasticity-aware price and promo optimization under competitive and operational constraints.
Selects and funds initiatives or projects to maximize benefit within skills, precedence, and budget windows.
Fairness-aware staffing and shift design respecting skills, preferences, targeted deliverables, and labor rules.
Unit commitment, fuel blending, and emission / cost trade-offs to meet demand at least cost within caps.
Multi-period capital planning — telco rollout, network expansion — to maximize coverage and ROI under QoS and budget constraints.
Instance right-sizing and commitment planning across providers to hit SLOs at minimum cost.
Keeps exposure and VaR within limits via Monte-Carlo plus optimization of positions and limits.
Optimizes cost-vs-carbon pathways; embeds emissions and policy targets directly in decisioning.
Space allocation, lease and retrofit timing, and energy optimization for built assets.
Browse the actual optimizers.
Every AOM lives inside a sector value chain. Open the sector page to see what runs where.