Honeywell Forge Asset Performance Management

The branding does resolve, with effort. Forge is a very broad Honeywell brand spanning buildings, cybersecurity, aerospace and industrial software, and the APM offering has been sold under several names. Honeywell’s own APM brochure, dated September 2024, settles it: its closing call to action directs readers to learn more about Honeywell Forge Performance+ for Industrials, Asset Performance, which is therefore the current marketed name, while the document itself is titled simply Asset Performance Management. The same brochure also names Asset Sentinel as the on-premises version of Honeywell’s asset performance product, which confirms the consolidation that third-party coverage had reported from the earlier Asset Sentinel, IR and IRIS products. Expect all of these names in proposals and references, and ask which underlying product and release is quoted.

Sensor requirement: no, and Honeywell states it twice over. The brochure describes a unified and vendor-agnostic platform allowing integration of data and custom models across different equipment types and OEM vendors into a single solution, and says it can connect the balance of plant assets across all sites and integrate with OEM condition-based monitoring systems. A 2025 Honeywell release on a North Sea deployment repeats the OEM-agnostic claim. Honeywell does sell optional sensors, Versatilis plug-and-play wireless sensors, for condition-based vibration monitoring, and it draws on Experion PKS Field Device Manager for instrument health and CCC control systems for turbomachinery, but none of that is a precondition. Notably, Corrosion Advisor is described as providing predictive corrosion insight without requiring corrosion sensors.

The mechanism is unusually well described, and it is a hybrid one rather than pure machine learning. Honeywell states it uses multiple modelling methods: first-principle-based efficiency models, machine-learning-powered anomaly detection algorithms, and mechanical fault monitoring as leading indicators, applying advanced analytics to a unified data set to recognise early signs of health issues and predict potential time to failure. It ships a standard asset model library of over forty asset types and more than one hundred asset variants, built by Honeywell engineers across reliability, instrumentation and controls, machinery, maintenance, process simulation and corrosion, and these include predefined fault models, fault trees, KPIs, causes, consequences and corrective actions. Honeywell also lists concrete failure modes it addresses, including boiler tube leakage, compressor lube oil system loss of pressure, furnace decoking cycle improvement, remaining useful life of a positive displacement pump and remaining corrosion allowance in crude distillation units. That is the level of specificity this category usually lacks.

Architecture is modular, with named Health Analytics and Performance Analytics submodules, a Configuration Studio, an Advanced Analytics Workbench, event management, performance case management and an explicit CMMS work order integration module. Deployment is on-premises or cloud-hosted via Honeywell Forge, and it can be rolled out incrementally from one site to an enterprise. Honeywell additionally offers subscription Remote Monitoring Services in which its own reliability experts own technical support, event monitoring and case management, so a managed wrapper is available. Honeywell publishes savings, earlier-detection and payback figures; none are recorded here.

Fits: large process operators, and Honeywell is explicit about which: it says APM is commonly deployed from plant to enterprise level across oil and gas companies, fuel refineries, petrochemical facilities, chemical plants, semiconductor manufacturing and mines. The hybrid modelling approach suits organisations with standard process equipment, compressors, pumps, heat exchangers, fired heaters and rotating machinery, where the shipped asset model library applies, and the first-principles efficiency models mean it addresses energy and performance degradation as well as mechanical health. It suits buyers who want vendor-supplied models with embedded domain knowledge rather than building their own, and those with genuinely mixed-OEM estates, given the vendor-agnostic design. Mining fleet operators have a specific named offering for mobile and fixed equipment.

Does not fit: buyers wanting to bring their own algorithms as the primary approach; Honeywell offers an Advanced Analytics Workbench but the value proposition is its own models. It is a weak fit for unusual or bespoke equipment outside the forty-plus asset types and hundred-plus variants in the library, since a model-led product is only as good as its coverage of your assets. Buyers outside heavy process industry and mining, and any small or mid-sized operation, are out of scope: there is no published price, no self-serve route and no trial, and Honeywell’s own material describes multi-site deployments with asset counts in the thousands.

Typical use cases: predictive health analytics combining pre-built fault symptom models with machine-learning anomaly detection; condition-based vibration monitoring using Versatilis wireless sensors; instrument health monitoring via Experion PKS Field Device Manager; predictive corrosion monitoring through Corrosion Advisor without corrosion sensors; first-principles performance monitoring to find efficiency degradation and its root causes, optionally with UniSim Design; turbomachinery insight from CCC control systems; health monitoring of mobile and fixed mining fleets.

Buyer caution: get the exact product name and release in writing. With Asset Sentinel, IR, IRIS, Forge APM and Forge Performance+ all in the lineage, and Asset Sentinel specifically being the on-premises variant, a proposal and a reference customer may not describe the same deployment. Also establish whether the Remote Monitoring Service is included or separately subscribed.