Dig Management & Tracking

Integrity digs are coordinated through spreadsheets, emails, and informal communication. Scheduling 200+ digs across multiple crews, rotating 21-on/4-off shifts, and two operating regions is manual, error-prone, and invisible to management.

Built a purpose-built dig tracking application with real-time status tracking (Site Selected → With CLEIR → CLEIR Approved → In Progress → NDE → Repair → Complete), crew scheduling with rotation awareness, and integrated Gantt views.

Management of Change — 5 Years in Maximo

MoC processes at pipeline companies are typically paper-heavy, slow to route, and hard to track. Changes to operating parameters, equipment, or procedures get bottlenecked. Compliance documentation is scattered.

Built a gated change management workflow in Maximo over 5 years as Technical Services Supervisor. Controlled changes to SCADA systems and Control Center operations. Each step must complete before the next begins. Structured, auditable, regulatory-ready.

Industry context: Formalized MoC processes can reduce recordable incident rates by up to 72% when fully adopted. — API Process Safety Guidelines

Alarm Management & Continuous Improvement

Merging pipeline operations creates an opportunity to align alarm philosophies across systems. Standardizing rationalization practices, implementing root cause analysis for recurring alarms, and building ongoing performance monitoring can improve operator response and reduce fatigue.

Responsible for alarm rationalization as Technical Services Supervisor for 5 years. 15+ years reviewing alarm trends on shift. Familiar with CSA Z662 requirements, priority discipline, and building sustainable review processes.

Industry context: Poor alarm management has been identified as a contributing factor in major process incidents, including one that resulted in $80M in damage and 26 injuries. — Yokogawa / ISA-18.2 case study

AI Leak Detection — Formal Proposal

In June 2024, proposed an AI-powered leak detection system to company leadership. Structured proposal covering cost-benefit analysis, liability assessment, resource planning, and technical implementation roadmap (CNN and RNN approaches, Azure integration).

"When we engage third-party vendors, we essentially hand over our valuable data, enabling them to build algorithms that they can then market to our competitors. The true value lies in the data itself."

The internal AI group reviewed the proposal and suggested building a working example.

Pipeline SCADA Simulator

Built a pipeline SCADA simulator to demonstrate the intersection of operational knowledge and technical capability:

• Validated hydraulic solver (WNTR — Sandia National Labs) with real pipeline physics
• Hazen-Williams friction modeling — same approach used in production systems
• Industry-standard HMI design practices (ISA-101 — grayscale base, color reserved for alarms)
• Pattern-based anomaly detection — identifies leaks, pump trips, valve restrictions
• Real-time monitoring via WebSocket with simulated SCADA data

CNN Corrosion Detection

Built a machine learning pipeline for corrosion detection on In-Line Inspection (ILI) data. Multiple detection models working together to identify and classify pipeline anomalies from inspection imagery. Integrated into a desktop application for field use.

Paid contract through CommandML — delivered to an ILI company.

LoRa Above Ground Marker (AGM)

Built a LoRa-based Above Ground Marker for detecting ILI tools (smart pigs) running underground. Used a geophone sensor to detect ground vibration from the passing tool and transmitted detection data over LoRa radio back to a local monitoring website.

Proof of concept — designed to solve a real field problem where pig tracking relies on personnel stationed along the pipeline.

AI-Controlled Robot

Built an autonomous vision-guided robot using Raspberry Pi and ESP32. Machine learning object detection for navigation and obstacle avoidance. Combines hardware engineering, embedded systems, and AI in a single project.

Phased Delivery

Large companies can't pivot overnight. Every solution in this roadmap follows a principle: earn the right to do the next thing by delivering the current one.

Start with tools that solve today's problems — dig tracking, MoC workflows. Build credibility with the people who use them daily. Then layer intelligence on that foundation — alarm rationalization, integrity-SCADA alignment. The more advanced capabilities (predictive analytics, AI-augmented leak detection) depend on the data foundation built in earlier phases.

Knowledge Transfer

When pipeline assets change hands, the biggest risk isn't technical — it's the institutional knowledge gap. Operational context about how systems work, why configurations were chosen, and what "normal" looks like on each pipeline segment often isn't documented.

Designing governance and knowledge capture processes to systematically document this context — operational databases, system-specific documentation, and training materials — reduces risk during transitions and builds a foundation for future decision-making.