☰

Contipro a.s.

One data model across fermentation, purification and nanofibers

Industry
Hyaluronan biotechnology and pharmaceutical APIs
Headquarters
Dolní Dobrouč, Czech Republic
Public information as of
January 2026

A4BEE prepared this analysis from publicly available sources. It reflects our own reading of Contipro a.s.'s published strategy and is not endorsed by, or produced in cooperation with, Contipro a.s.. Company website

Strategic priorities

Contipro produces sodium hyaluronate (HA) by bacterial fermentation at a single integrated site in Dolní Dobrouč, Czech Republic, serving pharmaceutical, medical-device and personal-care customers in more than 70 countries. About 50 percent of the workforce is in research and development, and the company holds over 120 patents with a sustained output of more than 15 high-impact publications a year. The 2024 financial year closed with net sales growth of 7.27 percent year-on-year and total assets up 10.08 percent, while net profit margin compressed by 0.93 percent on heavier research and modernization spend.

The production backbone has three distinct operation profiles that the company is trying to run on shared Industry 4.0 foundations. Fermentation for HA takes place in reactors up to 5 m³ and is governed by Process Analytical Technology (PAT) and closed-loop control described in the Facility of the Future program. Purification and downstream processing produce Ultra-Pure grades for intra-articular injections and ophthalmic applications certified to GMP, USP and EP. A separate, much smaller nanofiber line built on the 4SPIN electrospinning platform is at laboratory and pilot scale today and is targeted at tissue scaffolds and drug delivery.

On top of that runs an EU-funded research portfolio in regenerative medicine — IMMODGEL, DRIVE, NEURIMP, N2B-patch and AMCARE — that links HA biopolymers with advanced manufacturing methods including 3D bioprinting. The company has also begun operating in areas with stricter security expectations, including NIS2 in the European Union for OT environments, and serves pharmaceutical customers that operate under FDA and Japanese PMDA review windows alongside EU MDR.

Industry 4.0 work has produced a series of working pieces: an OPC UA (Open Platform Communications Unified Architecture) backbone for equipment communication, AI vision systems for foam detection on bioreactors, a YOLOv8-based microalgae cell classifier trained on microscope images, AR-assisted maintenance on the production floor and VR training modules for high-speed vial filling lines at 150 vials a minute. What the public record does not describe is a single integration layer that lets these pieces share data with the quality, regulatory and R&D sides of the business.

Challenges we see

  • Digital Integration

    Bridging legacy equipment into the connected production model

    Contipro runs Smart Fermentation with biosensors and IoT devices on the OPC UA backbone, but production-side assets installed across the lifetime of the facility include chillers, scales and sensors whose original control interfaces predate the current architecture.

    Where equipment sits outside the OPC UA model, the picture of the run lives next to the line rather than inside it, and integrating those points matters for any future end-to-end visibility work.

  • Operations Manufacturing

    Moving nanofiber output from laboratory scale toward commercial scale

    The 4SPIN platform covers a family of devices from multi-needleless emitters for higher throughput, through multi-jet capillary heads, to rotating and patterned drum collectors for aligned fibre sheets. Production transition targets tissue scaffolds, drug-delivery carriers and CD44-targeting HA micelles.

    Process validation, fibre-diameter consistency and biomedical-grade reproducibility become a larger component of the work as output moves from grams per batch to kilograms per batch.

  • Operations Operations

    Sourcing specialist talent in a rural Czech location

    Contipro is based in Dolní Dobrouč and competes for digital-native scientists and engineers with Prague and other central European technology hubs. About half of the workforce is in research and development, and the company runs an educational pyramid from a corporate kindergarten through university internships to build its local pipeline.

    Programmes that make day-to-day work more legible from a distance and that lift routine tasks off specialist desks change what the talent competition looks like for those hires.

  • Compliance Regulatory

    Filing for hyper-personalised HA therapeutics across multiple jurisdictions

    The strategic direction includes hyper-personalised HA-based drug-delivery systems for pathologies such as myocardial infarction, peripheral nerve damage and central-nervous-system delivery. The relevant filings span EU MDR for devices, FDA pathways and Japanese PMDA review windows.

    The documentation load rises with both the number of jurisdictions and the patient-specific variation in the underlying application, which means the review queue rather than the underlying science increasingly sets the calendar.

  • Compliance Regulatory

    Securing an expanding bioprocessing attack surface

    OPC UA, IoT sensors, cloud-integrated digital twins and AR equipment used in production all sit inside what is now classified as an OT environment under NIS2 in the European Union. Production takes place under GMP, and the same lines feed pharmaceutical API customers whose audit posture cascades back to Contipro.

    Each new connection added under the Industry 4.0 program changes what needs to be evidenced during a security review, so the security case has to keep up with the connection plan rather than be re-argued after each addition.

Opportunities, by urgency and business impact

Each bubble is one opportunity, numbered to match the list below. Further right means it bites sooner; higher means a bigger effect on the business. A bigger bubble means a bigger implementation effort.

Source: A4BEE analysis of public sources

  1. Reading production context from sensor through to scientist

    Industry 4.0 investments at Contipro — Smart Fermentation biosensors, OPC UA backbone, AI-vision foam detection, AR dashboards — run alongside production assets whose original control loops predate the digital programme and feed outputs to OT, IT, R&D and QC through separate paths.

    An ontology-based data layer with a single model of batch, run, parameter, assay and lot turns each of those paths into a single feed, so a query written once returns the same answer whether it comes from a fermentation engineer, a QC analyst or a regulatory writer.

    • Contipro 'about Contipro a.s.' page, CPHI Online listing
    • Contipro 'About European Hyaluronic Acid Manufacturer' page

  2. Closing the loop on bioreactor runs

    Contipro's fermentation control combines biosensors for dissolved oxygen and redox potential, PLC-driven feedback loops for aeration and AI vision for foam detection. The description of the Facility of the Future programme describes full 'lights-out' operation as the next stage rather than today's state.

    Tightening the loop between AI-vision inputs, autosampling and the controller extends the period a run can hold steady without operator intervention, which changes how often runs need human attention and how repeatable successive batches are.

    • Smart Fermentation Technologies review, MDPI, 2025
    • Patsnap Synapse, 'How AI is Revolutionizing Fermentation Process Optimization'

  3. Standardising the interface between modules and the orchestration layer

    Contipro has adopted the Module Type Package (MTP) standard, NAMUR 2658, to orchestrate bioreactors, chromatography systems, scales and other equipment without bespoke coding per machine, but the public account notes that proprietary subsystems are still in play.

    A documented MTP library with device adapters for the main equipment types means a new line can be assembled from existing objects, and HA molecular-weight variants or pilot nanofiber runs become configured on the same orchestration layer rather than re-engineered each time.

    • Contipro materials on MTP adoption and 'Plug & Produce' modularity
    • NAMUR NE 2658, Module Type Package standard

  4. Scaling 4SPIN nanofiber output against biomedical-grade criteria

    4SPIN is currently sold as a laboratory and pilot device family used inside Contipro's own programs and externally for R&D. Nanofibers from HA derivatives are being evaluated as tissue scaffolds, drug carriers and CD44-targeting micelles for cancer-cell delivery.

    Process modelling that predicts fibre diameter, alignment and resorption time from upstream parameters lets a scale-up candidate be assessed before a long wet-lab cycle, which shortens the path from a laboratory-scale trial to a reproducible biomedical-grade run.

    • 4SPIN product family documentation and electrospinning technical notes
    • Contipro nanofiber application portfolio, HA derivatives for drug delivery

  5. Carrying regulatory documentation through multi-jurisdiction filing cycles

    The therapeutics pipeline points toward hyper-personalised HA systems that cross EU MDR for devices, FDA review windows and Japanese PMDA filings, with Quality, R&D and Production contributing source data from separate systems.

    Narrow agents can assemble the first draft of a filing section from the underlying records, check the document against the template before it enters the review queue, and surface every controlled document a standards change touches, with a named reviewer approving each output.

    • Contipro regenerative medicine research portfolio (IMMODGEL, DRIVE, NEURIMP, N2B-patch, AMCARE)
    • EU MDR transition timelines and FDA / PMDA review expectations for HA-based combination products

What we'd propose

  • Enterprise AI

    An industrial data platform for fermentation, purification and nanofiber data

    An ontology-based data layer that defines batch, run, parameter, assay and lot once, then loads production data from OPC UA streams, QC instruments, R&D assays and nanofiber lines against the same model, with analytics and retrieval on top.

    • OPC UA backbone with documented address space

      Every published parameter has a contract

      Establish OPC UA as the secure machine-to-machine layer for the plant, with published information models for the bioreactor, chromatography and downstream equipment families so each value carries vendor-neutral meaning that follows it into the data layer.

    • Ontology for batch, run, parameter, assay and lot

      One agreed model of production

      Define the entities and relationships that describe a Contipro batch — vessel, run, parameter, sample, assay, lot — once, so OT, IT, QC and R&D systems load against the same model and a single query returns a comparable result across them.

    • Legacy equipment brought into the data path

      Existing assets reach the model

      Bring chillers, scales and other long-serving assets into the data path through vendor-neutral adapters and bridging devices, so the production view includes them without an equipment-replacement programme.

    • One feed replaces the many data extracts a fermentation, QC or regulatory team assembles today.
    • A 'golden batch' becomes a query against the model rather than a separate artefact.
    • New equipment and new HA grades attach to the same backbone instead of triggering another point-to-point integration.

    We've done this before

    See how we do Enterprise AI
  • Digital Lab

    Closed-loop bioreactor control with vision, autosampling and PAT

    A control stack that pairs the AI-vision foam detection already in place with autosampling, PAT instruments and ML-derived set points, so longer stretches of a fermentation run can hold their course without operator intervention.

    • Vision-based foam control

      Antifoam dosing from a camera

      Use external camera-based foam detection and adaptive dosing (pulse-width, vector control) to manage antifoam addition during the run, so the response is continuous rather than triggered manually.

    • Autosampling and PAT integration

      Results reach the controller

      Connect autosamplers to PAT instruments through a middleware layer so critical quality attributes — pH, dissolved oxygen, glucose, HA molecular weight — are measured and visible to the controller during the run.

    • Golden-batch ML advisory

      Comparisons during the run

      Train and deploy models that compare the current trajectory to historical best runs, surfacing deviations and parameter recommendations while the run is still on the line.

    • Longer stretches of a run hold their target without operator attendance.
    • Antifoam and feed dosing are evidence-graded by the system itself.
    • The same PAT data feeds release decisions instead of being re-measured at the end.

    We've done this before

    See how we do Digital Lab
  • Digital CDMO

    An MTP library for modular fermentation and downstream lines

    A reusable MTP (Module Type Package, NAMUR 2658) library with adapters for the equipment families at the Dolní Dobrouč site, so adding a new module — a new chromatography skid, an alternative bioreactor control head, a nanofiber line — does not mean a new integration project.

    • Object-oriented MTP module library

      Standardised service objects

      Develop a NAMUR 2658-aligned library of module objects with documented services for the bioreactor, chromatography and downstream equipment at the site, so adding a module is a configuration rather than a programming task.

    • Process orchestration layer

      One coordination plane

      Implement a process orchestration layer that coordinates MTP-compliant modules across a line, so HA molecular-weight variants, downstream steps and pilot nanofiber runs are configured against the same layer.

    • Vendor-agnostic device integration

      New vendors fit the same model

      Build AML (AutomationML) import paths and standardised adapters so equipment from new vendors joins the orchestration layer without bespoke code.

    • New production lines are configured from existing modules rather than wired from scratch.
    • Cross-system integration time drops because each module already speaks the same protocol.
    • Future equipment purchases are evaluated against a documented standard instead of restarting the integration discussion.

    We've done this before

    See how we do Digital CDMO
  • Enterprise AI

    Predictive digital twin for fermentation and 4SPIN scale-up

    A predictive process-model layer that uses historical fermentation and electrospinning data to forecast outcomes for new HA strains, reactor geometries and nanofiber recipes, so scale-up candidates can be triaged before long wet-lab validation cycles.

    • Fermentation scale-up model

      Predicting 5 m³ runs in advance

      Build simulation models that predict bacteria growth and HA yield at 5 m³ scale from laboratory parameters, so scale-up candidates can be ranked before a single production-scale run is committed.

    • Twin synchronisation with live runs

      Predictions stay in step

      Connect the twin to the live process data stream so the simulation updates against the running batch and drift between predicted and observed behaviour surfaces early.

    • 4SPIN nanofiber scale-up model

      Predicting nanofiber runs

      Model fibre diameter, alignment and resorption time as functions of upstream parameters so a new HA-derivative recipe can be compared to existing biomedical-grade runs before scaling the trial.

    • Fewer failed scale-up runs reach the production stage.
    • Time-to-grade decisions is set by the model rather than by the wet-lab queue.
    • Same simulation infrastructure serves HA fermentation and 4SPIN nanofiber scaling.

    We've done this before

    See how we do Enterprise AI
  • Agents

    AI agents for regulatory and quality documentation

    Narrow, reviewable agents that take the assembly part of regulatory and quality work — drafting deviation and change-control documents, mapping standards changes to affected documents, checking templates — off the specialist desk. A named reviewer approves each output.

    • Drafting from source records

      First drafts from the underlying data

      Generate first drafts of deviation reports, change-control forms and periodic review documents from the underlying electronic records in LIMS (Laboratory Information Management System), ELN (Electronic Lab Notebook) and MES (Manufacturing Execution System), so the author edits and judges rather than assembles.

    • Template and completeness checking

      Gaps found before the review queue

      Check each submitted document against its template and the site's own checklist before it enters the human review queue, returning missing or inconsistent sections as comments rather than as a later correction cycle.

    • Change-impact search across the document set

      Which documents a change touches

      When an EU MDR clause changes, a USP method is superseded or a customer's audit finding arrives, retrieve every controlled document that references the source and rank them by how directly each is affected.

    • Review queues move faster because documents arrive complete and referenceable.
    • The scope of a standards change is established by search rather than by recollection.
    • Each output is traceable to the source records it came from and signed off by a named reviewer.

    We've done this before

    See how we do Agents

Where QB Systems fits

Alongside our services we build QB Systems, hardware and software for bioprocess control. QB Systems is a product brand of A4BEE Sp. z o.o.

Digital maturity: today and target

Scored out of 100 across six dimensions. The target is what Contipro a.s.'s own published ambition implies — not a perfect score.

Source: A4BEE analysis of public sources
Data infrastructure 65 → 90
OPC UA backbone is in production and AI-vision and AR pieces are deployed, but legacy assets and document-bound workflows mean the same question still has to be re-asked across teams before it is answered.
Process automation 70 → 95
Fermentation feedback loops, foam vision and autosampling are running in pilots, while the Facility of the Future programme describes full closed-loop operation as the target rather than the current state.
IT/OT convergence 60 → 85
MTP adoption is underway but vendor-specific subsystems remain, and a documented module library with Process Orchestration Layer coverage is the next step the public materials point to.
Predictive analytics 55 → 85
Digital twins and YOLOv8 vision models are in use for monitoring, while scale-up simulation work for both fermentation and 4SPIN nanofibers is described as a future capability.
Workforce digitalisation 50 → 80
AR-assisted maintenance and VR training are running in pilots across the site, with broader operational rollout described as the route from pilot to enterprise use.
Cybersecurity posture 60 → 90
OPC UA backbone and Zero Trust principles are in place, and the production environment now sits inside NIS2 scope, which raises the bar for what needs to be evidenced during security reviews.

Check this yourself

Our Service Portal has free self-assessments and market comparisons. These are the ones that line up with what we've read above — no sales call required.

Think we've read this right?

Talk to us

Related reading

This is an independent analysis prepared by A4BEE from publicly available information as of January 2026. It reflects A4BEE's own interpretation and opinion, is not affiliated with, endorsed by, or verified with Contipro a.s., and may be incomplete or inaccurate. All company names and trademarks are the property of their respective owners. To request a correction or removal, contact [email protected].