AI Inside Maximo: How Predictive Maintenance, Condition Insights, and Agentic AI Are Reshaping Asset Management
AI Is Not Magic -- It Is Data and Algorithms
Matt Boehne, a recognized voice in the Maximo community, put it plainly in a 2026 presentation: "AI is rooted in two things -- data and algorithms. They're not magically figuring things out from nothing. They're working with the data we provide and with our knowledge of how things work."
This is the most important sentence anyone working with AI in Maximo will read. The platform's AI capabilities are real, they are powerful, and they are shipping in production today. But they are not magic. They depend entirely on the quality, consistency, and completeness of the data you feed them.
This article explains how AI actually works inside Maximo Application Suite 9 as of mid-2026. We cover the two prediction paths, the new Condition Insights agent, visual inspection, the AI Assistant, and -- most critically -- the data foundations that determine whether your AI initiative succeeds or fails.
The Two Prediction Paths in Maximo Predict
Maximo Predict works along two complementary paths. Understanding the difference between them is essential for setting realistic expectations and choosing the right approach for your assets.
Path 1: Failure History-Based Prediction
This path analyzes historical failure data to identify patterns. It ingests:
- Corrective maintenance work orders
- Failure codes and classifications
- Root cause analyses
- Repair frequency data
- Mean Time Between Failure (MTBF) patterns
The algorithm learns how assets have failed in the past and projects those patterns forward. This works especially well for repetitive assets with known failure modes -- pumps, motors, conveyors, HVAC components -- where failure history is well-documented and consistent.
The questions this path answers:
- Which assets are statistically most likely to fail next?
- Which failure patterns tend to repeat under similar operating conditions?
- What is the expected time to next failure based on historical intervals?
The limitation: this path is only as good as your failure data. If work orders lack failure codes, if root causes are not documented, or if MTBF data is incomplete, the predictions will be unreliable. This is the most common failure point in Maximo Predict implementations -- organizations expect AI insights but have not invested in the data discipline required to generate them.
Path 2: Condition and Sensor-Based Prediction
This path learns from real-time condition indicators:
- Vibration data
- Temperature readings
- Pressure measurements
- Runtime hours
- Degradation trends derived from asset health scores
Instead of asking "what failed before?", this approach asks "what behavior usually precedes a failure?" This is where early detection becomes possible -- identifying degradation patterns long before thresholds are crossed.
The questions this path answers:
- Which assets are showing early signs of degradation?
- What condition patterns preceded previous failures?
- How quickly is an asset's health declining?
The limitation: this path requires sensor data. If your assets are not instrumented, or if sensor data is not flowing into Maximo, this path is not available to you. The good news is that sensor instrumentation is becoming cheaper and more accessible. What required a six-figure capital project a decade ago can often be accomplished with off-the-shelf IoT sensors today.
How the Two Paths Work Together
The real power of Maximo Predict emerges when both paths operate simultaneously. A planner sees:
1. Rising failure probability from the history-based model
2. Declining health scores from the condition-based model
3. Known historical failure behavior for that asset class
This combination enables confident, explainable decisions. The planner is not blindly trusting a black-box prediction. They are seeing converging evidence from two independent analytical approaches, grounded in data they can inspect and understand.
As Stefan Hoffmann wrote in the IBM Community: "Predict does not tell you what will fail. It tells you where to look first." This is the right framing. AI in Maximo is a prioritization engine, not an oracle.
Maximo Condition Insights: Agentic AI Arrives
In December 2025, IBM announced Maximo Condition Insights, a significant evolution in how AI operates within the platform. Powered by IBM watsonx and agentic AI, Condition Insights delivers instant, explainable insights into asset health by analyzing multiple data sources simultaneously.
What Condition Insights Does
Condition Insights aggregates and analyzes:
- Work order history
- Time-series sensor data
- Meter readings
- Failure Mode and Effects Analysis (FMEA) records
- Alert history
- Asset specification data
The agent then produces:
- Trend and anomaly detection: Identifying patterns that individual data sources would miss
- Condition-to-failure-mode mapping: Connecting observed conditions to known failure modes from FMEA
- Corrective action recommendations: Suggesting specific maintenance actions based on the identified failure mode
- Conversational queries: Natural language interaction through the Maximo AI Assistant
Why Agentic AI Matters
Traditional AI in enterprise software is passive. It generates a score or an alert and waits for a human to act on it. Agentic AI is different. It actively reasons across data sources, forms conclusions, and recommends actions.
In the Maximo context, this means:
- Instead of a vibration alert that says "bearing temperature high," Condition Insights says "Bearing temperature on Pump P-101 is trending upward at 2.3 degrees per day, consistent with lubrication degradation failure mode FM-004. Recommended action: schedule lubrication replacement within 14 days."
- Instead of a work order backlog report, the AI Assistant answers "Which critical assets need attention this week?" with a prioritized list and the reasoning behind each recommendation.
The agent does not replace the reliability engineer. It amplifies them -- handling the data aggregation and pattern recognition that would take hours of manual analysis, and presenting findings in a form that enables faster, better decisions.
What Is Coming: Automated Work Order Creation
IBM has indicated that automated work order creation is on the Condition Insights roadmap. When this ships, the agent will not just recommend actions -- it will create the work orders, populate the job plans, and route them for approval. This closes the loop from detection to action without manual handoffs.
Visual Inspection: AI That Sees
Maximo Visual Inspection uses computer vision to analyze images and video from cameras, drones, and mobile devices. It detects defects, anomalies, and critical conditions that would be missed by human inspection or sensor-based monitoring alone.
How It Works
Visual Inspection models are trained on labeled images of assets in various conditions -- normal operation, early degradation, and failure states. Once trained, the model can:
- Identify corrosion, cracks, leaks, and other visible defects
- Detect missing or damaged components
- Verify proper assembly or installation
- Monitor for safety compliance (PPE, clearances, barriers)
The models can run at the edge (on a drone or mobile device) or in the cloud, depending on latency requirements and connectivity constraints.
Where Visual Inspection Excels
Visual inspection is particularly valuable for assets that are:
- Difficult to instrument: Overhead power lines, remote pipelines, offshore platforms
- Visually inspected today: Pressure vessels, storage tanks, structural steel
- Safety-critical: Where missed defects have catastrophic consequences
- High-volume: Where manual inspection cannot scale (thousands of solar panels, wind turbine blades)
The integration with Maximo means that when Visual Inspection detects an anomaly, it can automatically create a work order, attach the relevant image with the defect highlighted, and route it to the appropriate maintenance team. This eliminates the gap between inspection and action that plagues many manual inspection processes.
The Maximo AI Assistant: Natural Language for Operations
The Maximo AI Assistant enables natural language queries against Maximo data. Instead of navigating menus or writing SQL, users ask questions in plain language and receive structured answers.
What You Can Ask
- "Show me all critical assets that require maintenance this week"
- "What is the maintenance history for Pump P-101?"
- "Which work orders are overdue and assigned to my team?"
- "What is the current stock level for bearing part number BRG-440?"
- "Show me the failure trends for HVAC assets in Building 12"
How It Works Under the Hood
The AI Assistant translates natural language queries into API calls against the Maximo data layer. It understands the Maximo data model -- object structures, relationships, and business rules -- and constructs queries that respect security and data access controls.
This means a technician in the field can ask a question and get an answer without knowing which Maximo application to open, which fields to filter on, or how to construct a query. The assistant handles the translation.
The Data Dependency
The AI Assistant is only as good as the data it can access. If your asset registry is incomplete, if work order history is sparse, or if failure codes are inconsistently applied, the assistant's answers will reflect those gaps. This is not an AI limitation -- it is a data quality limitation that the AI makes visible.
The Data Foundation: Why Most AI Initiatives Struggle
Every Maximo AI capability -- Predict, Condition Insights, Visual Inspection, AI Assistant -- depends on data quality. This is the uncomfortable truth that separates successful AI implementations from failed ones.
The Most Common Data Gaps
Missing or inconsistent failure codes. If technicians close work orders without recording what failed and why, Predict has nothing to learn from. This is the single most common barrier to effective predictive maintenance. Incomplete asset hierarchies. If assets are not properly classified and related to their parent systems, the AI cannot understand failure propagation or aggregate insights at the right level. Sparse sensor data. Condition-based prediction requires continuous or near-continuous sensor data. Gaps in data collection create blind spots in the models. Uncalibrated meters. If meter readings are inaccurate, condition-based predictions will be inaccurate. Garbage in, garbage out applies with full force to AI.
The Mobilization Solution
The most effective path to better data is workforce mobilization -- putting the full power of Maximo in every technician's hands through mobile devices. When technicians can record failure codes, enter meter readings, and document findings at the point of work, data quality improves dramatically.
This is not just a technology deployment. It requires:
- Process discipline: Technicians must enter data consistently and completely
- Incentive alignment: Data entry must be seen as part of the job, not an administrative burden
- User experience: Mobile interfaces must be fast and intuitive, or adoptio
...
