From Rigid Schedules
to Resilient Operations.

Plans that survive contact with reality. In most process plants, the schedule the planner builds on Monday is broken by Tuesday — a delayed material, a quality hold, an unplanned breakdown, a rush order. Today, planners and supervisors absorb the disruption with phone calls, whiteboards, and overtime. Bodhee replaces that firefighting with a schedule that re-adapts in minutes against the same constraints a human would apply, and shows the planner why.

Yes. Bodhee orchestrates above MES and below planning. It consumes orders and master data from ERP (typically SAP S/4HANA or ECC, Oracle, or any other ERP), live execution events from MES and historian, quality results from LIMS, and asset events from CMMS. It then pushes back a feasible, optimised, re-adapting schedule that MES executes and ERP confirms.

APS (Advanced Planning & Scheduling) is built to optimise a plan once — typically weekly — against a snapshot of demand and capacity. Adaptive Scheduling is built to re-plan continuously against the live state of the plant. APS answers "what is the best plan for next week?"; Adaptive Scheduling answers "given everything that has changed in the last hour, what is the best plan for the next 24 hours, and which orders are now at risk?"

The Process Digital Twin (PDT) is the live representation of your plant that Bodhee reasons against — products, recipes, asset, materials, labour, quality holds, and the constraints that connect them. It is not a 3D model; it is a decision model. Every Bodhee recommendation traces back to a specific state of the PDT, which means every recommendation is explainable.

Bodhee Production Scheduling generates and continuously adapts a feasible, optimised production schedule against live constraints — orders, recipes, asset availability, materials, labour, and priority. When any of those change, the schedule re-plans within minutes with a goal to minimise the impact of that change on the schedule and the plant performance.

Demand / order data (campaigns, process orders, planned orders), products & recipes, assets, raw-material and intermediate availability, teams and operators and their skills, changeover and cleaning rules, and calendars (factory, shift, holiday). Most of these already exist in your ERP, MES, and master-data systems — Bodhee Production Scheduling reads them through standard connectors and the PDT layer.

All three, and the mixes of them that real plants run. Bodhee Production Scheduling treats each line on its own terms:

• Campaign-mode lines — orders are grouped and sequenced so similar products run back-to-back, amortising changeover and cleaning across the campaign and minimising cross-contamination risk.
• Batch-mode lines — orders are scheduled batch-by-batch against recipe, asset compatibility, and batch duration, with cleaning, validation, and hold-time windows respected between batches.
• Continuous-mode lines — production is scheduled as a throughput rate against asset constraints, run-length limits, and the upstream / downstream lines that feed or are fed by the continuous train.

A single plant can run all three side by side — the scheduler does not force one model onto every line. Where one mode's output feeds another (e.g., a continuous reactor feeding a batch crystalliser feeding a packaging campaign), Bodhee schedules the handoffs as constraints rather than treating each line in isolation.

Cleaning is modelled as a schedulable process in its own right — with its own tasks, durations, resources, and dependencies — not as a fixed overhead bolted onto production. Cleaning tasks link to production batches through the trigger patterns that real plants actually run on:

• Pre-production — cleaning before the first batch or campaign starts.
• Post-batch / post-campaign — cleaning at the end of a batch or at the end of a campaign.
• Mid-batch — cleaning triggered after a specific task inside a batch completes.
• Production-blocked-on-cleaning — a production task that cannot start until a specified cleaning process is complete.

Changeovers follow the same model — typed by product-pair, costed, and sequenced by recipe family — and validation holds (post-cleaning verification, pre-processing steps) are modelled as their own gated task class with their own duration and downstream blocking effect. Bodhee Production Scheduling respects every cleaning, changeover, and validation rule, and sequences orders to minimise their cumulative cost when the order book allows.

Yes — overrides are first-class, not exceptions. A planner can fix or freeze an order, lock an equipment to an order, exclude an asset, exclude a set of operators, set priority on certain materials, or force a window; the scheduler adapts the rest of the plan around the override.

Minutes for typical single-plant scope; configurable for larger horizons and multi-site scope. Re-plan cadence is event-driven by default — the system triggers when a constraint materially changes, not on a fixed clock. Typically granular rescheduling (with operator re-allocation) runs for the next 24–48 hours, and strategic re-planning runs for the next 2–4 weeks, depending on the level of disruption and the user-configured cadence.

Both. The default is a single recommended plan, ready to dispatch. The planner can also run what-if scenarios — "what if we lose Line 2 for six hours", "what if we accept the rush order", "what if we push the changeover to night shift" — and compare KPI impact side-by-side before committing.

Yes. The scheduler consumes material status from ERP and MES, detects shortages or late arrivals against the planned sequence, and re-sequences orders to absorb the disruption rather than break the plan. Where re-sequencing is not enough, it surfaces the OTIF risk early with the trade-offs spelled out. Material availability data from ERP is enriched using the active schedule for forecasted material availability (e.g., semi-finished-goods availability for finished goods).

Yes. Multi-site deployments are supported with site-level autonomy as the default — each plant runs its own schedule with its own constraints — and cross-site orchestration where the order book genuinely flows between sites. Bodhee supports both topologies.

No. APS systems plan at weekly / monthly cadence against forecast and capacity assumptions. Bodhee adapts at daily / shift cadence against live plant state. They are complementary: APS sets the boundary conditions and the production targets; Bodhee adapts the schedule inside those boundaries as the plant runs.

No. MES executes and records work on the shop floor; Bodhee schedules what the MES will execute. Most deployments leave MES untouched and integrate Bodhee above it.