Falkonry Time Series AI

On ownership, the honest answer is that the position is unclear and should be checked with the vendor. IFS announced in August 2023 that it had signed an agreement to acquire Falkonry, with completion expected in the fourth quarter of 2023, and the stated rationale was adding anomaly detection to IFS ERP, EAM, FSM and ESM products. However, no completion announcement could be verified, and falkonry.com today carries no IFS branding of any kind, markets the product independently, and offers its own trial. Aggregator records also show funding activity dated after the announced acquisition, which is inconsistent with a completed takeover. The possibilities include a deal that did not close, a technology acquisition that left the brand operating separately, or a subsequent separation. This profile records the product as active and independently marketed, because that is what the vendor site shows, and flags ownership as unverified. A buyer should ask Falkonry directly who owns the company and what the relationship with IFS is, and anyone considering IFS Cloud should ask IFS the same question from the other side.

The current product is Time Series AI, positioned within what Falkonry calls an Operational Resilience Data Platform. It ingests, normalises, aligns, enriches, analyses and visualises time series data, and provides pattern recognition, anomaly detection and no-code or low-code tooling aimed at engineers rather than data scientists. The positioning language, operational resilience, sovereign and air-gapped deployment, suggests a defence and critical infrastructure orientation alongside industry.

Sensor requirement: no, and the connectivity is specific. Falkonry describes itself as sensor-agnostic and working with your current telemetry, and names connection to existing SCADA and DCS systems and PLC controllers over Modbus and OPC UA. Naming actual protocols rather than claiming open connectivity is exactly the kind of statement that can be checked, and it puts this product among the better-evidenced in the category on integration. Falkonry also offers its own physics sensors, including high-rate power quality measurement, as an option rather than a requirement. A historian is not explicitly mentioned, which is worth confirming if that is your primary data source.

Deployment is the strongest aspect: edge, cloud, on-premises, air-gapped and sovereign are all offered, which is a wider range than any other product in this category and genuinely useful for defence, utilities and isolated OT environments. Falkonry also offers a free one-week offline trial using historical Parquet data, which is a rare and practical way to assess a product of this kind without an integration project. Anomaly detection is recorded as supported because it is a stated core function with a described data pipeline; failure prediction is recorded as unknown, because pattern recognition and anomaly detection are described while a prediction model is not. No specific third-party product integrations are named and no pricing is published.

Fits: engineering teams that have time series data from SCADA, DCS or PLCs and want to find patterns and anomalies in it themselves, without a data science team and without shipping data to a vendor cloud. The no-code and low-code tooling is explicitly aimed at engineers, and the trial on historical Parquet files means a reliability engineer can test the premise on real data in a week. The deployment range makes it one of very few credible options for air-gapped, sovereign or defence environments, and for utilities and critical infrastructure with strict data egress rules.

Does not fit: buyers who need a packaged asset health and reliability suite with asset hierarchies, FMEA libraries and prescriptive work orders, none of which is described. This is a time series analytics platform, not an APM suite, and a maintenance organisation expecting the latter will be disappointed. It is also not for organisations that need contractual certainty about the vendor, given the unresolved ownership position, or for teams that want named out-of-the-box integrations to specific CMMS or ERP products, since none are published.

Typical use cases: finding operational anomalies in historian or SCADA time series without building a data science pipeline; running analytics at the edge or inside an air-gapped network where cloud platforms are prohibited; letting process and reliability engineers label and recognise patterns themselves; assessing whether predictive value exists in your data at all, using the offline trial, before committing to a larger APM programme.

Buyer caution: resolve the ownership question in writing before contracting, confirm historian connectivity if that is your main source, and be clear that you are buying an analytics platform rather than a maintenance management system.