☰

Enifer Oy

Intelligence for continuous fermentation, operations for a lean workforce

Industry
Mycoprotein Fermentation
Headquarters
Espoo, Finland
Public information as of
January 2026

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

Strategic priorities

Enifer is commercialising the PEKILO® mycoprotein fermentation platform — a continuous process that runs 2,000 hours or more without a batch boundary — for aquafeed and pet food markets. The €33M Kantvik facility in Kirkkonummi, Finland (construction 2024, ramp-up planned for early 2026) targets 3,000 tonnes per annum. The company raised a €15M Series B in May 2024 led by Taaleri Bioindustry Fund I, following an €11M Series A in April 2023 and €24M in public and grant funding, totalling approximately €50M raised.

The operational challenge is distinctive: a 15-person shift crew runs a 24/7 continuous process inside a brownfield industrial site, alongside integration with Siemens PCS neo (new automation), Nordzucker legacy site utilities, and a Brazil-based FS Bioenergia data environment. The biology — maintaining a fungal steady state over thousands of hours — does not pause for handovers or manual inspection rounds.

Enifer's business model combines its own production with technology licensing of the PEKILO® process to ethanol producers globally. The pilot with FS Bioenergia in Brazil (500 tpa, scaling to 10,000 tpa) is the first deployment outside Finland. Both tracks require the same underlying data infrastructure: a unified view of process parameters, lab results, and feedstock composition that travels as well as the biological process does.

Challenges we see

  • Bioprocess Operational Technology

    Detecting metabolic drift in a continuous fermentation run before it causes a culture crash

    Enifer's PEKILO® process runs 2,000+ hours continuously, maintaining a fungal steady state in the Paecilomyces variotii organism. Unlike batch fermentation, where a contamination event ruins one run, a continuous process crash requires a full system shutdown, sterilisation, and a slow re-establishment of biomass equilibrium — a multi-day recovery with no product in that time.

    At 2,000-plus hours of continuous operation, the window between a subtle metabolic drift and a full culture crash is measured in hours. Off-gas analysis (CO2 and O2 evolution rates) is the earliest real-time signal of that drift, but only when it is read by a model that knows what the run's normal profile looks like — not by operators watching a screen.

  • Supply Chain Quality Control

    Compensating for feedstock variability without stopping a continuous run

    Enifer uses industrial side streams as feedstock: lactose permeate from Valio in Finland, corn thin stillage from FS Bioenergia in Brazil. By definition, side stream composition varies with upstream production schedules and seasonal harvest. The fungus requires a specific Carbon-to-Nitrogen ratio; uncontrolled variability causes yield drops or overflow metabolism.

    Where feedstock composition changes continuously and the process runs without interruption, manual sampling and adjustment cycles are too slow to maintain steady-state conditions. Inline sensors that read composition at the intake and a feed-forward control loop that adjusts nutrient dosing before the process reacts are what keeps yield within specification through a variable feedstock window.

  • Operations Labor

    Covering physical inspection requirements across a 15-person shift roster

    The Kantvik plant is designed for approximately 15 employees and 3-4 people per shift, operating 24/7. The site houses 12-metre fermenters, extensive piping, and hundreds of valves across a brownfield industrial precinct. A minimal shift crew keeps cost competitive but cannot physically inspect every asset on every round.

    At this workforce scale, inspection coverage has to be directed by data rather than by routine. A predictive maintenance algorithm that ranks inspection priority by equipment health scores — derived from historian data — directs the available people to the places that most need attention, rather than spreading effort across all locations equally.

  • Digital Integration IT/OT Convergence

    Connecting Siemens PCS neo to Nordzucker utilities, laboratory systems, and the Brazil partner

    Despite selecting modern Siemens PCS neo as the distributed control system, Enifer's Kantvik site sits within the Nordzucker industrial precinct and inherits legacy utility systems. The Brazil FS Bioenergia partnership runs on a separate SAP and Coupa stack. Lab QC data — HPLC, protein analysis — sits in disconnected systems from the process historian.

    Where process data, lab results, and partner data live in separate systems, the correlation analysis that identifies the conditions behind a high-yield run cannot be run at all. Bringing them into one model — with agreed data structures at each boundary — is the prerequisite for both operational optimisation and the batch-data compilation that EFSA and FDA submissions require.

  • Compliance Regulatory

    Compiling EFSA and FDA regulatory submissions from fragmented data sources

    Enifer is simultaneously pursuing EFSA Novel Food approval and has received FDA GRAS status. Both require proving manufacturing consistency across production batches. Data currently sits in Excel (lab), Siemens PCS neo (process), and partner emails (feedstock), making dossier compilation a manual exercise prone to transcription errors.

    When a regulatory submission is built from data assembled by hand, each batch report is a project. When the same data is generated as a continuous by-product of production — with an audit trail from instrument to historian to report — the submission is a retrieval rather than an assembly exercise.

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. Predictive analytics for metabolic drift in continuous fermentation

    Continuous fermentation requires maintaining biological steady state over 2,000+ hours. Operators watching a screen cannot detect subtle metabolic drifts (respiratory quotient shifts, morphology changes) fast enough to prevent culture crashes that require multi-day recovery periods.

    Deploy machine learning models analysing off-gas data (CO2 and O2 evolution rates), dissolved oxygen, and pH to predict culture drift 4-8 hours before it reaches critical thresholds, enabling micro-adjustments in feed rates to sustain steady-state conditions.

    • The company's CEO, interviews 2024-2025
    • Enifer Kantvik facility digital context, VTT spin-off documentation

  2. A unified industrial data platform connecting Kantvik, Nordzucker utilities, and Brazil

    Critical data handshakes between Siemens PCS neo, Nordzucker legacy utilities, FS Bioenergia's SAP environment, and disconnected laboratory systems are manual or absent. Lab QC data sits apart from the process historian, preventing Golden Batch correlation analysis.

    Architect a cloud-based industrial data platform that ingests data from all process, lab, and partner sources through agreed data structures at each boundary, providing a single source of truth for operations and a structured evidence base for regulatory submissions.

    • Enifer Kantvik brownfield integration requirements, 2024
    • FS Bioenergia Brazil partnership data bridge requirements

  3. A PEKILO digital twin as a technology transfer and licensing tool

    Enifer's licensing business model requires transferring the PEKILO® process to ethanol producers globally. PDF manuals cannot capture the process knowledge needed to replicate performance with different feedstocks and in different physical configurations.

    Package the validated Kantvik process model into a deployable digital twin that FS Bioenergia and future licensees can simulate, use for operator training, and run what-if scenarios against — creating a recurring SaaS revenue stream from the process IP.

    • The company's CEO, on process transfer to Brazil with different feedstocks
    • Enifer technology licensing model, investor presentations

  4. Remote operations and predictive maintenance routing for the lean workforce

    A 3-4 person shift cannot provide 24/7 expert-level process oversight, and a skeleton crew cannot physically inspect all critical assets on every round. Complex biological or mechanical issues occurring overnight wait until the Lead Bioprocess Engineer arrives.

    Use Siemens PCS neo's web-based architecture to build a secure remote command capability, and implement predictive maintenance algorithms directing the available crew to specific high-risk inspection points based on equipment health analytics rather than uniform patrol routes.

    • The Lead Automation Engineer, on web-based PCS neo remote accessibility
    • Enifer Kantvik operational design, 15-person workforce rationale

  5. Real-time carbon accounting and automated batch report generation

    EUR20M+ in public funding is conditionally tied to environmental performance metrics. Investors and customers require auditable proof of carbon displacement per batch, but current reporting is retrospective and assembled by hand from fragmented data sources.

    Integrate energy meters and mass balance data into a live sustainability analytics layer, and connect the historian and LIMS to a validated reporting engine that auto-populates EFSA and FDA compliant batch reports as a continuous by-product of production.

    • Finnish Climate Fund and EU NextGenerationEU funding terms
    • Taaleri Bioindustry Fund I investment mandate, May 2024

What we'd propose

  • Digital CDMO

    AI-driven bioprocess intelligence for continuous fermentation

    We deploy soft sensor capabilities and machine learning models analysing off-gas composition, dissolved oxygen, pH, and morphology indicators to predict culture drift 4-8 hours before it reaches a critical threshold, enabling operators to make micro-adjustments that preserve steady-state conditions without stopping the run.

    • Predictive culture analytics

      Metabolic drift detected before it crashes

      Build ML models that establish the normal off-gas profile for a given feedstock composition and run phase, then continuously compare the live signal against that profile to issue alerts when drift trajectories cross prediction thresholds — giving operators hours of notice rather than minutes.

    • Feedstock compensation with inline NIR

      Composition read at intake, dosing adjusted automatically

      Integrate inline near-infrared spectroscopy at the feedstock intake manifold to read C:N ratio and key substrate concentrations in real time, then route those readings into a feed-forward control loop that adjusts nutrient dosing before the process biology reacts to the change.

    • Golden batch benchmarking

      Every run measured against the best run

      Correlate current run parameters against the historical best-performing batches to generate a real-time deviation score, guiding operators toward the conditions that have historically produced the highest yield rather than toward a generic setpoint.

    • Culture crash recovery time — which is days of no production — is avoided by catching drift before it propagates.
    • Feedstock flexibility is increased because the process compensates for variability rather than being derailed by it.
    • The model improves over time as more run data accumulates, making early drift detection progressively more accurate.

    We've done this before

    See how we do Digital CDMO
  • Enterprise AI

    Unified industrial data platform for Kantvik and Brazil

    We architect a cloud-based data platform that brings Siemens PCS neo process data, Nordzucker site utility data, FS Bioenergia's Brazil production data, and laboratory results into one ontology-driven model — a single source of truth for operations, regulatory submissions, and technology transfer.

    • MTP-standardised equipment connectivity

      All equipment speaks one language

      Implement Module Type Package (MTP) interfaces for all process equipment connected to Siemens PCS neo, ensuring that new equipment additions are configured rather than engineered from scratch, and that partner systems can connect through the same standardised information model.

    • Brazil-Finland data bridge

      Quality verification from Espoo without travelling

      Build a secure cloud connector that ingests real-time process data from FS Bioenergia's production unit and makes it visible in the same model as the Kantvik operations — so Espoo HQ can verify quality performance without waiting for email attachments.

    • LIMS and SCADA integration for batch genealogy

      Lab results linked to the batch record

      Connect the laboratory information management system and HPLC, GC-MS, and protein analysis instruments directly to the process historian through a validated middleware layer, so that each batch record carries its associated lab results automatically.

    • The Golden Batch analysis that is impossible across disconnected systems becomes straightforward in a unified model.
    • Regulatory submissions for EFSA and FDA are compiled by retrieval rather than assembled by hand.
    • The same data platform that supports Kantvik operations is the infrastructure on which the Brazil technology transfer runs.

    We've done this before

    See how we do Enterprise AI
  • Enterprise AI

    PEKILO digital twin for global technology licensing

    We package the validated PEKILO® fermentation process — reaction kinetics, morphology dynamics, feedstock response curves — into a deployable digital twin that licensees can use for operator training, what-if scenario testing, and process performance simulation before committing to physical infrastructure.

    • Virtual process simulation

      Fermentation dynamics in a browser

      Build a calibrated model of PEKILO® fermentation that replicates the specific mass transfer, heat transfer, and morphology behaviour of the Kantvik reference process, allowing licensees to explore process windows and understand the impact of different feedstocks before physical trials.

    • Operator training environment

      Practice without production risk

      Create an immersive training environment where new operators can manage process alarms, respond to drift scenarios, and practice recovery procedures in a simulation — reducing startup errors when a new licensee goes live.

    • Licensed deployment with IP protection

      Process knowledge protected in encrypted models

      Deliver the digital twin as a hosted or on-premises model with usage tracking and encrypted parameter sets, so that Enifer's core process IP is protected while still enabling licensees to run their own optimisation within the licensed envelope.

    • The licensing model becomes a recurring SaaS revenue stream rather than a one-time know-how transfer.
    • Licensee startup errors are reduced by giving operators practice time in a simulation before the physical plant runs.
    • Process consistency across geographies is maintained because every licensee starts from the same digital reference.

    We've done this before

    See how we do Enterprise AI
  • Digital CDMO

    Remote operations centre and predictive maintenance routing

    We build a secure remote monitoring and command capability on top of Siemens PCS neo's web-based architecture, enabling the Lead Bioprocess Engineer to troubleshoot from any location, and we implement risk-directed inspection routing so the 3-4 person shift crew directs their limited time to the assets that most need attention.

    • Web-based remote HMI access

      Plant visibility from anywhere

      Configure Siemens PCS neo's web-based interface for secure remote access, giving the Espoo team and the off-shift Lead Bioprocess Engineer full alarm visibility, trend analysis, and override capability from any location — without needing to be on site to manage a developing situation.

    • AR-guided maintenance workflows

      Expert guidance for night-shift technicians

      Integrate an assisted-reality workflow layer (RealWear or equivalent) so that a night-shift technician can share a live feed with the off-site Lead Bioprocess Engineer and receive step-by-step visual guidance through a complex mechanical or process intervention.

    • Predictive maintenance routing

      Inspection directed by equipment health

      Build an equipment health scoring model from historian data — vibration signatures, valve cycle counts, temperature trending — and generate a daily inspection priority list that directs the available crew to the specific high-risk inspection points rather than requiring a complete physical patrol of the entire site.

    • Complex biological or mechanical issues at 3 AM no longer wait for the Lead Bioprocess Engineer to travel to site.
    • The 15-person organisation operates with the diagnostic awareness of a team three times its size.
    • Inspection effort is concentrated on the assets that most need it, rather than spread evenly across all assets regardless of condition.

    We've done this before

    See how we do Digital CDMO
  • Digital Lab

    Automated regulatory reporting and real-time carbon accounting

    We connect energy meters, mass balance data, historian, and LIMS into a validated reporting engine that auto-populates batch reports for EFSA Novel Food and FDA GRAS submissions, and produces a live sustainability dashboard for investors and lenders — all as a continuous by-product of production rather than a retrospective assembly exercise.

    • Validated batch report generation

      Reports written by the process

      Configure a reporting engine that pulls structured data directly from the process historian and LIMS, mapping each data point to the required regulatory template fields so that batch reports are generated automatically at the close of each production period rather than compiled by hand.

    • Real-time carbon accounting per batch

      Every batch has a carbon passport

      Integrate energy meters and mass balance calculations into a live sustainability analytics layer that computes specific energy consumption, Scope 1 and 2 emissions, and feedstock carbon intensity per production batch — producing a digital sustainability passport that accompanies each batch to offtake partners.

    • Investor and lender live dashboard

      Taaleri and Finnish Climate Fund see live KPIs

      Expose a secured URL providing Taaleri Bioindustry Fund I, Finnish Climate Fund, and other investors with real-time visibility into carbon performance, overall equipment effectiveness, yield, and energy intensity — replacing quarterly PDF reports with continuous transparency.

    • Regulatory submissions for EFSA and FDA are retrievals from a structured database rather than manual assemblies.
    • The €20M in conditional public funding is protected because carbon performance evidence is generated continuously rather than estimated retrospectively.
    • Investor confidence is supported by live operational transparency rather than by periodic reports that investors receive weeks after the period closes.

    We've done this before

    See how we do Digital Lab

Digital maturity: today and target

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

Source: A4BEE analysis of public sources
Process automation 55 → 90
Siemens PCS neo is deployed at Kantvik and Insta is the system integrator, representing a current-generation automation platform. The gap is in the predictive analytics layer — without ML-based metabolic drift detection, the automation platform manages setpoints but does not anticipate the biological events that cause crashes.
Data integrity 40 → 95
Lab QC data (HPLC, protein analysis) is disconnected from the process historian, Nordzucker legacy utilities data is separate from Enifer's systems, and FS Bioenergia Brazil data is accessed by email. No unified data backbone exists, making Golden Batch correlation and regulatory batch report assembly manual work.
Lab connectivity 35 → 85
No modern LIMS is documented in place; lab data moves in Excel. The 21 CFR Part 11 compliance gap for electronic records is structural, not merely a workflow convenience issue — regulatory inspectors for EFSA and FDA will require evidence chains that manual processes cannot reliably produce.
Digital twin capability 25 → 80
No digital process model of the PEKILO fermentation is documented as operational. The digital twin work — which is central to the licensing business model — has not yet been built, and the Sweco BIM models of the Kantvik facility represent a starting point rather than a process simulation.
Remote monitoring 30 → 85
PCS neo's web-based architecture was selected partly for remote accessibility, but a structured remote command capability — with role-based access, alarm escalation workflows, and AR-assisted maintenance — has not been built out. The Brazil data gap means Espoo has no live visibility into the FS Bioenergia process.
Supply chain integration 40 → 80
Valio feedstock scheduling is managed by email and phone; FS Bioenergia data handoff is via email attachments. Visibility into upstream feedstock composition — which directly affects fermentation yield — is not instrumented, making real-time feedstock compensation impossible.

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 Enifer Oy, 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].