Data Engineer/Scientist
Must be located in Massachusetts/New Hampshire or Rhode Island
1. Project Overview
The Global Service organization seeks to engage a qualified Data Engineer or Data Scientist to develop advanced reliability “life‑ing” models using historical ServiceMax labor, parts, and work order data. This initiative will support global reliability, service strategy, parts planning, and product quality improvement by providing data‑driven reliability estimates, early‑life failure identification, and usage‑based reliability insights across key systems and components.
2. Objectives
The consultant will deliver a data foundation and predictive modeling framework that enables the organization to:
- Quantify usage‑based life information (time and pulses where appropriate) for major subsystems and parts.
- Identify early‑life failure patterns using pulses, cycles, time‑in‑service, and work order data.
- Develop global reliability models for systems and components based on historical field performance.
- Create a scalable, repeatable modeling process that can be adopted by internal teams after engagement
3. Scope of Work
3.1 Data Engineering & Preparation
- Extract, clean, and transform historical data from ServiceMax, including:
- Work orders
- Failure codes
- Parts consumption
- Asset attributes
- Usage signals (pulses, cycles, hours, etc.)
- Develop a unified reliability dataset suitable for statistical modeling.
- Identify data gaps, inconsistencies, and required remediation.
3.2 Reliability & Life Modeling
- Develop Weibull, survival analysis, and time‑to‑failure models.
- Build usage‑based reliability curves for components and systems.
- Generate statistical outputs including:
- B10/B50 life estimates
- Early‑life failure rates
- Failure distribution curves
- Reliability trends by geography, product family, and usage profile
- Validate models for accuracy, repeatability, and business relevance.
3.3 Predictive Model Development
- Build predictive algorithms estimating probability of failure based on real‑world usage.
- Develop early-warning indicators for specific components or systems.
- Provide machine learning or regression‑based forecasting where appropriate.
3.4 Deliverables & Handover
- Clean, structured reliability dataset derived from ServiceMax history.
- Usage‑based B10, Weibull life models for designated parts and systems.
- Early‑life failure detection framework using pulses/time/work order data.
- Documentation package, including:
- Data dictionary
- Modeling methodology
- Assumptions and validation steps
- Repeatable code or notebooks (Python/SQL)
- Final presentation summarizing insights, reliability findings, and recommendations.
Never miss an opportunity! Create an alert based on the job you applied for.
