Predictive Maintenance Dashboard - Connected Operations Lab Module

Predictive Maintenance Dashboard - Connected Operations Lab Module

Predictive Maintenance Dashboard - Connected Operations Lab Module

A simple dashboard that turns raw, messy equipment signals into clear, risk-based maintenance insights.

A simple dashboard that turns raw, messy equipment signals into clear, risk-based maintenance insights.

Overview

This dashboard consumes simulated fleet telemetry, applies simple maintenance-risk scoring logic, and visualizes the output in a customer-ready interface. Instead of chasing complexity or ML hype, the focus is on clarity: showing the path from signal → risk → action the way a Sales Engineer would explain it in discovery, demos, or value conversations.

The pattern mirrors workflows used across fleet tech, equipment monitoring, manufacturing, and industrial IoT:

  • ingest time-series data

  • surface anomalies and emerging risks

  • prioritize assets that require attention

  • present information in a way that makes decisions obvious

How It Works

Data Simulation (Python)

Generates multi-vehicle telemetry streams containing metrics such as:

  • coolant temperature

  • intake air temperature

  • engine RPM

  • vibration score

  • derived maintenance-risk indicators

Simulates realistic variation across assets to model the noise SEs deal with in real operational data.

Scoring Logic

Applies simple, explainable rules to categorize maintenance risk:

  • thresholds for normal vs concerning values

  • per-asset scores rolled into Low / Medium / High buckets

  • contribution breakdown by signal

This keeps the workflow transparent and easy to walk through with non-technical stakeholders.

Dashboard (Streamlit)

Displays:

  • top-risk assets

  • trendlines and outlier spikes

  • driven-reasoning (“which signals contributed most to the score”)

  • drill-down into individual assets

The UI completes the workflow by making the technical story frictionless to follow.

Tech Stack & Business Value

Tech:
Python, Streamlit, Pandas, Matplotlib

Why This Matters in a Sales Engineering Context:

This module demonstrates the exact type of narrative SEs deliver around connected systems: how scattered sensor values become structured insights, and how insights become decisions. It’s not just a dashboard; it’s a miniature version of what IoT, fleet, and industrial SaaS platforms do under the hood.

For hiring teams, it showcases:

  • Ability to simplify and explain data-flow logic

  • Comfort with time-series patterns and operational signals

  • Skill in turning raw data into a coherent business story

  • Familiarity with failure points (noise, spikes, drift, bad data)

  • Proof you can walk a customer through “here’s the signal → here’s the risk → here’s the next step”


This is the workflow customers actually buy, not the tech.

Risk Scoring

Risk Scoring

Anomaly Detection

Anomaly Detection

Fleet Overview

Fleet Overview

Prioritized List

Prioritized List

Trend Analysis

Trend Analysis

Score Breakdown

Score Breakdown

Streamlit UI

Streamlit UI

Explainable Logic

Explainable Logic

Demo Links

GitHub Repo
Demo Loom: COMING SOON

© 2025 Drake Wildes.

All rights reserved.

Technical Sales Engineer Portfolio