An industry doesn't run in isolated modules. We built a platform that doesn't either.
How Alyva connected demand, materials, capacity, suppliers, warehouses, transport, production, quality, machines, maintenance and outcome into a single industrial architecture, with MRP II at the center of the decision.
Alyva was born from a simple observation: planning materials is only part of the industrial problem.
A factory needs to know what to produce, how much, when, what to buy, where each material is, how it arrives, where it will be stored, which resource processes it, whether there is capacity, whether the batch is released, whether the machine will be operational and whether the deadline is still possible. All of this happens at the same time. The system needed to work the same way.
↦ The scenario
The operation was connected physically. The information was not.
An industry is a network of dependencies. Sales changes demand, demand changes planning, planning changes purchasing, purchasing changes receipts, inventory changes production, machines change capacity, quality releases or holds, maintenance pulls a machine out of the plan, transport determines when something arrives. And every decision produces financial consequences.
Even so, most systems treat these areas as independent compartments: ERP on one side, production planning in spreadsheets, inventory on another screen, carriers outside the system, quality on forms, maintenance in separate software, and planning trying to pull it all together.
The problem wasn't a lack of software. It was a lack of continuity between decisions.
↦ The problem
MRP is not a calculation screen. It is a chain of decisions.
The question seemed simple: how much do we need to buy or produce? But it unfolded fast.
- →Is there finished product? If not, how much to manufacture, and which components does that consume?
- →How much is in stock, reserved, held or in transit? What is the minimum batch and the lead time?
- →Is there a machine and available capacity? Is maintenance scheduled? What sequence is possible?
- →Did the material pass quality? In which warehouse and location is it? How long does it take to arrive?
- →And once produced, how does the product reach the customer?
Answering material need alone wouldn't solve it. We had to build around the concept of MRP II: planning connected to capacity and to operational reality.
↦ The idea
If one decision changes the whole chain, the system needs to understand the cascade effect.
Alyva was not conceived as a collection of administrative modules. The architecture was designed so that each domain shares context with the others. The central logic became a single flow:
While another flow happens at the same time:
And all of these events feed Costs, Fiscal, Finance and Analytics. The system stops asking only "what happened?" and starts answering: "what does this change trigger across the rest of the operation?"
↦ The solution
Alyva: a platform where planning and execution talk to each other.
Fronts that work separately, but gain value when they operate connected.
From demand to net requirement
Explodes the multilevel BOM and deducts stock, reservations, transit, losses, batches and lead times. Not just how much is missing, but how much will be missing and when.
The delivery date becomes a chain
Works backwards, from delivery to purchasing, turning a commercial date into operational dates and revealing delays before the line.
Having material is not being able to produce
Connects materials to work centers, routings, times, availability and maintenance. This is where Alyva enters MRP II territory.
Planning becomes action
Converts requirement into a supply flow, with suppliers, lead times, terms and performance. Purchasing sees what will be missing before the rupture.
Knowing how much exists is not knowing where it is
Receiving, multiple warehouses, addressing, picking, packing, FEFO and traceability. Physical stock is not operational availability.
The chain doesn't end at the gate
Inbound and outbound transport: carriers, routes, freight, scheduling, tracking and CT-e. A finished product is not a delivered order.
The order becomes operational context
Connects BOM, routing, resources, dates and quality, and returns real consumption, losses, yield and times. The plan gets feedback from execution.
The machine talks to planning
Industrial execution or integration with an existing MES, with bidirectional communication of orders, status, consumption and quality.
A common language
Layered integration for a heterogeneous plant: OPC UA, MQTT, Modbus, Profinet, EtherNet/IP, MTConnect. The Edge Gateway collects and normalizes even without a connection.
The indicator explains the context
Availability, performance and quality tied to equipment, order, product, shift and maintenance. Not just "did OEE drop?", but "what was happening when it dropped?".
Capacity depends on asset health
Condition, hours, alerts and orders join the production context (CMMS integration), with predictive strategies by vibration and temperature.
Producing does not mean releasing
Inspection, specification, in-process control, non-conformities, hold and release, connected to suppliers, production, batches and stock (LIMS integration).
From raw material to customer, and back
Multidirectional traceability across products, batches, suppliers, equipment and processes, shrinking the investigation universe in recall and audit.
The process controls the system
A Dynamic Business Rules Engine (conditions, decisions, actions): operational policy becomes configuration, not a new version of software.
Process is data too
Approvals adaptable by value, risk and deadline, sequential or parallel. The system follows the organization's process, not the other way around.
Evidence is born with the operation
Quality, safety, environment and audit in the flow (ISO 9001, IATF 16949, GMP, FDA 21 CFR Part 11, ANVISA). Compliance stops being a later reconstruction.
↦ The cascade effect
Everything can start with a single order.
A customer confirms demand. Stock is checked, there is not enough quantity.
MRP calculates the requirement. The BOM is exploded. Some components are available, some reserved, some in inspection, some need to be purchased.
Lead times are applied. Purchase requirements are generated. Inbound transport enters the calculation.
The materials arrive. Receiving records them, quality inspects, the WMS addresses them, the material is released.
Capacity is checked. One machine has scheduled maintenance. The sequence has to change. The orders are rescheduled.
Production starts. Equipment sends data. Real consumption updates stock. Quality tracks the parameters.
The product is finished and the batch released. The WMS receives the finished goods. Picking and packing prepare the order. Shipping creates the transport.
The product ships. Tracking returns events. Fiscal processes the documents. Finance receives the effects. The dashboards reflect the new reality.
It all started with an order. That is the difference between a set of modules and a connected industrial system.
↦ The engineering principle
Data should be born where the process happens.
The goal was never to build dozens of screens for someone to manually reconstruct the reality of the factory. The architecture aims to make each operational event generate context automatically.
A receipt updates stock. An inspection changes availability. A movement updates the position in the warehouse. A production run consumes materials. A machine reports its state. A non-conformity holds a batch. A maintenance job changes capacity. A sale changes demand. A transport updates the estimated arrival.
MRP receives the new reality and recalculates the future. The operation starts producing its own decision context.
↦ The capability that stayed
From transactional ERP to an industrial operation platform.
The main capability is not in a specific screen. It is in the integration of decisions. The architecture brings them together, not as independent software, but as parts of the same chain:
MRP and MRP II
Multilevel BOM, net requirements, lead time, capacity, routings, sequencing and scenario simulation.
From supplier to the line
Suppliers, purchasing, requisitions, receiving and inbound logistics.
Where it is and in what condition
Inventory Control, multiple warehouses, addressing, batches and expiry, RFID, barcode, FEFO, picking and packing.
All the way to the customer
Carriers, routes, freight, scheduling, transport orders, tracking, fiscal integration and CT-e.
Execution with feedback
Orders, routings, reporting, consumption, yield, losses and work centers.
Release or hold
Inspections, control plans, non-conformities, deviations, holds and releases, LIMS integration.
The health of capacity
Equipment, maintenance, condition, availability and CMMS integration.
The connected shop floor
MES, SCADA, IoT, Edge, OPC UA, MQTT, Modbus, Profinet, EtherNet/IP, MTConnect and OEE.
Rule and evidence
Workflows, dynamic rules, audit, traceability and industry compliance.
From process to result
Billing, fiscal, finance, APIs, external integrations, dashboards and analytics.
↦ The result
From fragmented information to a connected operation.
Alyva turned a software problem into a process engineering problem. The question stopped being "which module do we need to build?" and became "what information does this decision need to receive, and who needs to know when it changes?"
Without inventing ROI numbers not yet measured, the result we can state is structural: an architecture able to follow the industrial flow from demand to delivery and return execution data back to planning. From demand to the shop floor. From the shop floor to delivery. From execution back to planning.