PDEL®
Performance Dependency Elucidation Language
What Is PDEL®?
PDEL® is a computational language for causal modeling—enabling engineers to map system performance back to root physical, operational, and organizational drivers in a mathematically rigorous way.
The Problem
Traditional process models simulate equipment behavior but cannot explain why performance degrades. Statistical models identify correlations but cannot distinguish causation. Organizational analysis points to human factors but cannot quantify their impact on business outcomes.
PDEL® bridges these gaps by providing a structured syntax for expressing causal relationships that span physics, operations, and business value.
Core Principles
Causal Clarity
Every performance metric must be traceable to its physical or organizational causes—not just correlated variables.
Computational Rigor
Causal relationships are expressed mathematically, enabling quantitative scenario exploration rather than qualitative judgment.
Cross-Domain Integration
The language links physics (degradation mechanisms), operations (maintenance strategies), and finance (cost/revenue impacts) in unified models.
Organizational Traceability
Human decisions and process gaps are modeled as system inputs—making organizational factors quantifiable.
How It Works
PDEL® provides a structured syntax for defining:
- Performance Variables: What you're trying to optimize (uptime, efficiency, throughput, EBITDA)
- Dependency Maps: Which input variables directly influence each performance variable
- Functional Relationships: Mathematical expressions for how inputs affect outputs (linear, exponential, probabilistic)
- Uncertainty Quantification: Probability distributions for inputs with inherent variability
- Scenario Definitions: Structured ways to explore "what if" questions across multiple variables
Example Application
Heat Exchanger Fouling → EBITDA Impact
Using PDEL® to model a refinery heat exchanger:
- Physical Layer: Fouling rate = f(feed sulfur content, temperature, velocity)
- Operational Layer: Cleaning frequency = f(fouling rate, maintenance budget, turnaround schedule)
- Performance Layer: Heat transfer efficiency = f(fouling thickness, cleaning frequency)
- Business Layer: Production capacity = f(heat transfer efficiency); EBITDA = f(production, energy cost, cleaning cost)
The model reveals that feed sulfur variability—a procurement decision—has quantifiable EBITDA impact through its effect on fouling rates. This enables cross-functional optimization: procurement can evaluate feed contracts against maintenance cost and production value.
Why It Matters
PDEL® transforms vague organizational challenges into engineering problems with quantifiable solutions. Instead of saying "we need better communication between operations and maintenance," you can say "reducing the delay in failure notification by 24 hours reduces unplanned downtime by 8%, worth $2.3M annually."
This precision enables executives to make resource allocation decisions with confidence—because the causal chain from action to outcome is explicit and testable.