Honeywell Forge Cognition is a building and industrial operations platform for owners, facility managers and engineering teams responsible for complex physical assets. In a commercial real estate setting, its immediate target is the control room workflow around HVAC alarms, operating anomalies and proposed changes to building systems. Rather than asking an operator to investigate every alert manually, the platform applies AI agents to connected operational data, develops a probable explanation and can use simulation to examine a proposed response. Honeywell positions the product as a step from conventional automation toward supervised autonomy.
The distinction matters because Forge Cognition is not a broad property management system. It does not replace lease administration, accounting, resident communications or work order platforms. Its value sits deeper in the operational technology layer, where building automation systems, equipment signals and engineering procedures meet. That makes its relevance strongest for campuses, data centers, hospitals and portfolios with substantial mechanical complexity. A conventional office owner with outsourced engineering and limited system access may find the deployment burden harder to justify.
Honeywell says a campus pilot reduced unnecessary alerts and identified potential energy savings. Those claims are directionally encouraging, but buyers should separate demonstrated workflow improvement from projected financial impact. Identified savings are not the same as measured savings appearing on utility bills, and Honeywell does not publish enough standardized results to establish a dependable portfolio benchmark. Analysis: the product should therefore be evaluated as an operational decision system with promising early evidence, not as a proven energy savings contract.
The strongest reason to consider Forge Cognition is the combination of Honeywell’s building controls background with agentic AI and simulation technology developed using technology from NVIDIA and Dell Technologies. The principal caution is commercial and evidentiary transparency. Public pricing, standardized implementation requirements, customer counts and broad outcome data are not published. BestCRE would place Forge Cognition on a shortlist for technically sophisticated owners, but only through a tightly scoped deployment with agreed baselines, human approval controls and measurable operating outcomes.
What Honeywell Forge Cognition Does and How It Works
Forge Cognition connects to supported building and industrial systems and interprets operational information through specialized AI agents. For a facility team, the practical workflow begins with alarms, equipment behavior or a question about changing operations. The system can gather relevant signals, reason across the available context and present a proposed diagnosis or action. Honeywell describes an approach in which the technology learns from operations, proposes improvements and acts only with customer approval. That approval boundary is important in buildings where an incorrect command could affect comfort, critical environments or equipment life.
Simulation is the more distinctive component. Instead of treating every recommendation as an instruction to be executed immediately, the platform can test operating changes in a modeled environment. A facilities team might use that capability to examine the likely effects of a control adjustment before applying it to supported systems. The quality of the result will depend on the completeness of the connected data, the fidelity of the model and the condition of sensors and controls. Forge Cognition cannot compensate automatically for missing points, incorrect equipment mappings or persistent calibration errors. Buyers should require Honeywell to document which inputs drive each important recommendation.
The product also aims to reduce the investigative burden associated with high alarm volumes. Honeywell has reported fewer unnecessary alerts in a campus pilot, suggesting that the platform can help distinguish actionable conditions from operational noise. The vendor has also moved toward deployments that customers can scope, which should make a bounded evaluation more practical than an enterprise commitment. Analysis: a sensible first deployment would cover one building type, a defined HVAC subsystem and a small set of recurring alarm classes. The buyer could then compare investigation time, validated alarms, comfort outcomes and realized energy consumption against an agreed baseline before expanding. Public information does not establish that Forge Cognition supports every building automation vendor or every legacy protocol, so compatibility must be confirmed during technical discovery.
9AI Framework: Dimension by Dimension Analysis
CRE Relevance: 7/10
Forge Cognition earns a 7 for CRE relevance because it addresses a costly and persistent property operations problem: turning HVAC alarms and building system data into prioritized engineering decisions. That is directly useful to owners and operators of complex assets, especially campuses and facilities with large mechanical plants. The limitation is breadth. This is not a complete property management application, and it does not address leasing, accounting, tenant engagement or general portfolio administration. Its practical relevance also depends on whether a property has compatible systems, accessible operational data and an internal team capable of acting on recommendations. Analysis: the product is highly relevant within the building operations layer but only moderately relevant to the wider CRE organization. Buyers should classify it as specialized operational intelligence, not as the central system for managing a property portfolio.
Data Quality and Sources: 7/10
Forge Cognition receives a 7 for data quality. Honeywell has deep familiarity with building controls and mission critical operating environments, which provides a credible foundation for interpreting equipment signals. The platform is also designed to work with data from physical systems rather than relying only on documents or generic language models. However, the vendor does not publish a detailed, independently validated account of data coverage, model training sources, sensor quality controls or performance across varied CRE portfolios. Building data is frequently incomplete, inconsistently named and affected by faulty sensors. Those conditions can weaken even a sophisticated model. Analysis: data quality should be treated as deployment specific. A buyer should test point mapping, timestamps, equipment relationships, missing data handling and calibration before relying on recommendations. Strong source systems may produce strong results, while neglected controls infrastructure will constrain the platform.
Ease of Adoption: 6/10
Forge Cognition scores 6 for ease of adoption. The ability to scope a deployment gives buyers a practical route to start with a defined operating problem instead of committing an entire portfolio. The agent interface may also reduce the effort required to interpret alarms once integrations and models are working. Adoption is still unlikely to resemble activating a conventional cloud application. The product must connect with physical building systems, reflect site specific equipment and preserve appropriate approval controls. That requires coordination among facilities staff, information technology, cybersecurity teams, controls contractors and Honeywell. Training is also necessary so operators understand when to accept, reject or investigate an AI recommendation. Analysis: sophisticated organizations can manage this process, but smaller owners may lack the staff and data governance needed for a successful rollout. Buyers should ask for a written commissioning plan, responsibility matrix and acceptance criteria before signing.
Output Accuracy: 6/10
Forge Cognition earns 6 for output accuracy because its simulation and supervised recommendation model offer sensible safeguards, but public evidence remains limited. Honeywell reports fewer unnecessary alerts in a campus pilot, which indicates that the system may improve prioritization. The company has also described potential energy savings, yet potential savings do not prove that recommended changes delivered lasting reductions under normal operating conditions. No broad public benchmark establishes diagnostic precision, false negative rates or performance across different building types. Analysis: accuracy should be evaluated by use case rather than through a single headline measure. Buyers should create a labeled sample of historical alarms, compare the platform’s conclusions with engineer findings and review both missed issues and unnecessary escalations. Recommendations affecting critical environments should remain subject to explicit human approval until site specific performance is established. Simulation results should also be checked against actual post-change behavior.
Integration and Workflow Fit: 7/10
Forge Cognition receives a 7 for integration because it is designed for connection to operational systems and comes from a vendor with an established presence in building automation. Its value depends on reading real equipment and control data, so integration is central to the product rather than an optional extension. Honeywell also developed the platform using technology from NVIDIA and Dell Technologies, supporting deployments that combine AI computation with operational environments. The unresolved issue is interoperability. Honeywell does not publish a complete matrix showing every supported building automation platform, protocol, data model or third-party system. Analysis: customers with Honeywell controls may encounter a clearer path than portfolios assembled from several control vendors and generations. Before procurement, owners should inventory systems by property, confirm read and write capabilities, determine where computation occurs and document responsibility for ongoing connector maintenance. Cybersecurity review must cover access privileges, network segmentation and approval controls.
Pricing Transparency: 3/10
Forge Cognition scores 3 for pricing transparency. Honeywell does not publish subscription prices, implementation charges, hardware requirements or standard service packages for the product. That prevents buyers from estimating total cost without entering a sales and technical scoping process. The absence of public pricing is understandable for a system whose cost may vary by site, equipment count, integrations and computing architecture, but it still makes early comparison difficult. Analysis: buyers should request a commercial proposal that separates software, hardware, integration, data preparation, commissioning, training, support and future expansion. The proposal should also identify recurring expenses and any usage based charges. A scoped deployment should have a fixed definition of success and a clear ceiling on services cost. Without those protections, a technically promising pilot can become an open-ended integration project whose economics are difficult to evaluate against Facilio, Diald, Willow or targeted controls optimization services.
Support and Reliability: 7/10
Forge Cognition earns 7 for support based on Honeywell’s established role in building and industrial operations, where implementation and service capabilities are often as important as the software. Buyers may benefit from access to teams familiar with controls, equipment behavior and operational technology. That background is particularly valuable when recommendations touch systems that affect occupant comfort or critical operations. However, Honeywell does not publish Forge Cognition specific service levels, response times, staffing ratios or a detailed support package. The actual experience may vary by contract, region and the parties responsible for existing controls. Analysis: procurement teams should not assume that support for the AI layer is included with traditional building service agreements. Contracts should identify escalation paths, coverage hours, responsibility for model issues, connector maintenance and procedures for disabling recommendations. A named technical owner and scheduled performance reviews would reduce ambiguity during the first deployment.
Innovation and Roadmap: 8/10
Forge Cognition scores 8 for innovation. Combining specialized AI agents with simulation and physical building systems is a meaningful advance beyond dashboards that merely display faults or summarize trends. The ability to evaluate a proposed operational change before execution could help engineering teams make faster decisions while retaining human control. Honeywell’s use of NVIDIA and Dell Technologies also signals an architecture intended for computationally demanding physical AI applications. Innovation does not by itself guarantee operating value. The platform must still prove that its models reflect actual equipment, that recommendations survive real world constraints and that simulations remain current as buildings change. Analysis: Forge Cognition is among the more technically ambitious offerings in CRE operations, but it should be purchased for validated workflows rather than for the agentic AI label. The strongest evaluation will compare it with both modern operations platforms and simpler fault detection tools on measurable site outcomes.
Market Reputation: 8/10
Forge Cognition receives 8 for market reputation because Honeywell is a publicly traded, established provider of building automation and industrial technology. Its presence in mission critical environments gives the product institutional credibility that a young software vendor may not have. Owners evaluating a long term operational platform can reasonably view Honeywell as a counterparty with substantial domain experience. The product itself is newer than the parent company, so Honeywell’s corporate reputation should not be mistaken for extensive Forge Cognition adoption. The vendor does not publish a customer count, retention rate or a large set of detailed CRE case studies for this specific offering. Analysis: the appropriate score recognizes a strong sponsor while reserving judgment on the product’s own market record. Reference calls should focus on installations with comparable building types, control systems and staffing models, not on unrelated Honeywell products or industrial deployments.
Who Should Use Honeywell Forge Cognition
Forge Cognition is best suited to institutional owners, corporate real estate groups and facility operators managing mechanically complex properties. Campuses, data centers, hospitals, laboratories and large mixed portfolios are the clearest candidates because alarm volume, equipment interactions and operating consequences can justify deeper analysis. A suitable buyer will have accessible building system data, experienced engineers and enough recurring operational issues to measure improvement. Organizations already using Honeywell controls may have a more direct technical path, although compatibility and project scope still require confirmation.
The strongest buyer profile is an operations leader who can define a narrow first problem and supply credible baseline data. That could mean reducing time spent reviewing recurring HVAC alarms, testing specific control changes or improving the consistency of fault triage across several similar buildings. Information technology and cybersecurity teams must participate early because the system touches operational networks and potentially proposes physical actions. Analysis: the product becomes more attractive when avoided investigation time, comfort performance and verified consumption changes can be measured together. It is less compelling when the business case depends only on a broad promise of autonomy.
Who Should Not Use Honeywell Forge Cognition
Small landlords, lightly instrumented buildings and organizations seeking a general property management suite should not make Forge Cognition their first technology purchase. The platform does not replace accounting, leasing, work orders or tenant communication. Properties with poor sensor coverage, undocumented controls or limited access to building data should address those foundations before adding an AI decision layer. Otherwise, implementation effort may exceed the value of the recommendations.
Buyers that require transparent list pricing, rapid self-service deployment or extensive public proof across comparable CRE portfolios should also wait. Honeywell does not publish pricing or enough standardized performance evidence to support a simple desktop purchase decision. Analysis: organizations without an internal facilities champion are especially likely to struggle, because evaluating recommendations requires operational judgment. A conventional fault detection platform, controls recommissioning project or managed engineering service may provide a clearer first step for these teams.
Pricing and ROI Analysis
Honeywell does not publish pricing for Forge Cognition. Public information does not specify license structure, minimum contract value, hardware charges, implementation fees or whether pricing is based on buildings, equipment, data points, users or computing consumption. Buyers should expect a scoped proposal shaped by system complexity and deployment architecture. The commercial comparison should use total cost over the planned evaluation period, including integration, data preparation, commissioning, training, cybersecurity review, ongoing support and any required computing equipment.
Analysis: ROI should be calculated from verified operating changes, not identified opportunities. Suppose a buyer establishes its own annual baseline for alarm investigation labor, avoidable service visits and energy expense within the pilot boundary. The defensible benefit is the portion actually reduced after adjusting for weather, occupancy and other operational changes. Divide that verified annual benefit by the full annualized cost of software, hardware and services to estimate the benefit to cost ratio. The buyer should also calculate payback by dividing initial deployment cost by verified monthly benefit. No universal savings assumption is appropriate because Honeywell has not published enough comparable outcome data.
Integration and CRE Tech Stack Fit
Forge Cognition belongs beside the building automation system, fault detection tools and operational data infrastructure, not beside the accounting ledger. Its useful inputs are equipment signals, alarms, operating states and contextual relationships among physical assets. Outputs should flow into existing engineering processes, including review, approval and work execution. Buyers should clarify whether recommendations can create records in a computerized maintenance management system or property operations platform, because public documentation does not establish a complete catalog of CRE application connectors.
Portfolio fit depends on system diversity. A Honeywell centered environment may offer a more coherent starting point, while mixed vendor portfolios need property level compatibility testing. Owners should confirm supported protocols, data latency, historical data requirements, write permissions and deployment location. Analysis: start with read access and human approved actions, then consider greater automation only after performance is validated. Integration governance should also define who maintains equipment mappings when controls, tenants or operating sequences change.
Competitive Landscape
Facilio is the most relevant alternative for buyers seeking a broader connected operations platform across maintenance, energy and portfolio workflows. It may fit organizations that want operational coordination beyond AI based alarm analysis. Diald is a more focused option for teams interested in AI assisted building engineering and fault interpretation. Willow competes through digital twins and contextualized asset data, which can be attractive when the broader goal is a navigable digital representation of buildings and infrastructure. Each alternative still requires scrutiny around integrations, deployment services and verified outcomes.
Within BestCRE’s broader property management and operations category, established platforms such as AppFolio, Entrata, DoorLoop and Yardi address substantially different jobs. They are stronger candidates for property accounting, leasing and administrative workflows, but they are not direct substitutes for physical system simulation and HVAC alarm triage. VTS is more relevant to leasing and asset management, while Moved concentrates on resident moving services. Analysis: comparing Forge Cognition directly with those platforms would obscure the buying decision. The appropriate shortlist is Forge Cognition against Facilio, Diald, Willow, the owner’s existing fault detection system and a controls optimization engagement. Honeywell is differentiated by its building controls heritage and simulation oriented approach, but it must prove incremental value over tools already receiving the same equipment data.
The Bottom Line
BestCRE recommends a controlled evaluation of Honeywell Forge Cognition for owners with complex HVAC operations, accessible system data and a capable engineering team. Do not begin with a portfolio wide commitment. Choose a bounded subsystem, establish baseline performance and require the platform to demonstrate better alarm prioritization, defensible recommendations and measurable operating benefit under human supervision.
The product’s technical direction is credible, and Honeywell brings meaningful building operations experience. The purchase case remains constrained by unpublished pricing, incomplete interoperability detail and limited public outcome evidence for this specific platform. Analysis: proceed when the vendor accepts clear success criteria, transparent total cost and an exit point after the initial scope. Wait if the proposed business case rests mainly on potential savings or if existing controls data has not been audited.
About BestCRE
BestCRE.com is the definitive authority on commercial real estate AI, analysis, and investment intelligence. Every article advances the platform’s mission to help CRE professionals identify, evaluate, and adopt the best tools and strategies in the industry. We benchmark platforms using the 9AI Framework so CRE leaders can compare tools with clear evidence. Explore the category map at 20 CRE sectors for deeper coverage across the CRE stack.
Frequently Asked Questions
What is Honeywell Forge Cognition?
Honeywell Forge Cognition is an agentic AI and simulation platform for physical operations in buildings, campuses and industrial environments. For commercial real estate teams, its central use is interpreting information from supported building systems, triaging HVAC alarms and evaluating potential operating changes. It is intended for building owners, facility managers and operations teams, particularly those responsible for complex mechanical systems. It is not a conventional property management suite and does not replace lease administration, accounting or tenant communication software. Honeywell developed the platform using technology from NVIDIA and Dell Technologies. Analysis: buyers should view it as a specialized decision layer for operational technology, with value determined by system compatibility, data quality and the ability of facility staff to validate and act on its recommendations.
How does Honeywell Forge Cognition compare with Facilio, Diald and Willow?
Forge Cognition differentiates itself through Honeywell’s controls experience, AI agents and simulation of proposed operational changes. Facilio generally presents a broader connected operations proposition spanning maintenance, energy and portfolio workflows. Diald is a focused alternative for AI assisted building engineering and fault analysis. Willow approaches the problem through digital twins and contextualized asset information. The right choice depends on whether the buyer prioritizes deep interaction with physical operations, broad workflow management or an asset centered digital representation. Analysis: Forge Cognition should win only if a pilot shows that its simulation and alarm triage deliver incremental value beyond the owner’s existing controls and fault detection tools. Buyers should compare the same buildings, equipment classes, integration costs and outcome measures across vendors.
How much does Honeywell Forge Cognition cost?
Honeywell does not publish pricing for Forge Cognition. The vendor also does not publicly specify a standard subscription structure, implementation fee, hardware package or minimum deployment size. Cost is likely to depend on the selected facilities, supported systems, integration work, data preparation and computing architecture, but buyers should obtain those details directly rather than assume a pricing model. Analysis: request an itemized proposal separating software, hardware, commissioning, training, support and expansion costs. Calculate ROI only from benefits verified after deployment, such as reduced investigation labor, avoided service activity or measured consumption changes. Potential savings identified by the platform should not be counted as realized value. A fixed pilot scope and a defined cost ceiling will make comparisons with other vendors more credible.
Who should use Honeywell Forge Cognition?
Forge Cognition is most appropriate for institutional building owners, corporate facilities groups and operators of complex sites such as campuses, data centers, hospitals and laboratories. These environments tend to have large HVAC systems, significant alarm traffic and engineering teams capable of reviewing AI recommendations. Suitable buyers should already have accessible building data, clear operational ownership and cybersecurity processes for connected systems. Organizations with Honeywell controls may have a logical starting point, although compatibility should still be documented. Analysis: the ideal customer can isolate a recurring problem, establish a credible baseline and measure results over a defined period. Small properties, lightly instrumented buildings and owners seeking accounting or leasing software are unlikely to receive enough value to justify the deployment effort.
Will Honeywell Forge Cognition eventually operate buildings autonomously?
Honeywell presents Forge Cognition as part of a shift from automation toward autonomy, but its current positioning emphasizes recommendations and actions taken with customer approval. That is the appropriate model for CRE environments where changes can affect comfort, equipment, safety or critical operations. Greater autonomy is technically plausible as models, integrations and operating evidence improve, but Honeywell does not publish a universal timetable for fully autonomous building operation. Analysis: owners should evaluate autonomy as a sequence of controlled permissions. Begin with observation and recommendations, progress to simulated testing and human approved execution, then consider narrowly defined automatic actions only after site specific accuracy is established. Contracts and system design should preserve audit trails, access controls, rollback procedures and the ability for facility staff to override the platform.
Related Reviews
Compare Honeywell Forge Cognition with other Property Management & Operations tools BestCRE has scored: DoorLoop (73/100), Entrata (72/100), Equiem (71/100) and Moved (66/100). Browse every review in CRE Property Management & Operations, the full Best CRE AI Tools directory, or the 20 CRE Sectors hub.
