Portrait of Joel Baarbé

Chemical Engineer · Process Control & Automation

Joel Baarbé

Building the automation, real-time data infrastructure, and estimation algorithms behind continuous biomanufacturing at McMaster University.

Get in touch See my work

About

From plant floor to process model.

Joel smiling outdoors with mountains and evergreen trees behind him

I’m a chemical engineer specializing in advanced process control, industrial automation, and real-time process data infrastructure. I like work that spans the whole stack: wiring a PLC into a SCADA system, building the database that stores its data, and writing the model and estimator that turn that data into decisions.

I’m currently an M.A.Sc. candidate in Chemical Engineering at McMaster University under Dr. Prashant Mhaskar. My thesis is on spectroscopy-driven concentration estimation for continuous monoclonal antibody manufacturing. Working with Sartorius Stedim Biotech, I’ve helped develop, integrate, and operate the upstream and downstream automation and connectivity systems for a continuous mAb process. I also served as the downstream (DSP) lead during an integrated manufacturing campaign.

Before graduate school I earned my B.Eng. in Chemical Engineering at McMaster on the Dean’s Honour List, specializing in Process Systems Engineering. During my internship at Nylene Canada I validated plant modifications that saved roughly 2–3% and implemented real-time wastewater monitoring.

Outside the lab I run races from 10 km up to the full marathon, make music, and help lead a summer camp. That’s where I once wrote an optimization model to assign electives to 800+ campers.

  • 250+valves controlled by a SCADA system I built
  • 8,000+process variables streamed every 2 seconds
  • ~20×more spectral wavelengths captured for modelling
  • 3.85graduate cGPA at McMaster

Research & Focus Areas

What I work on

Spectroscopy-Driven Estimation for Continuous Antibody Manufacturing

Continuous biomanufacturing runs nonstop, so operators need to know protein concentrations as the process runs, not hours later from an offline sample. My M.A.Sc. thesis combines inline UV and Raman spectroscopy with multivariate modelling (PCA/PLS) and mechanistic state estimation to infer concentration in real time across a continuous monoclonal antibody process. That is the foundation for tighter monitoring, process analytical technology (PAT), and advanced control.

Current research · Dr. P. Mhaskar

Advanced Process Control

Model predictive control (MPC), PID tuning, and digital control design that keep processes stable, safe, and on target, even through disturbances.

State Estimation

Extended Kalman and particle filters that combine noisy measurements with process models to estimate states no sensor can see directly.

Process Modelling

Mechanistic ODE-based and dynamic models, calibrated to experimental data and written in state-space form so they can be used for estimation and control.

Multivariate Statistics

PCA and PLS for high-dimensional process and spectral data, turning thousands of correlated signals into models you can understand.

Optimization

LP, NLP, and MIP formulations in Pyomo and GAMS for equipment design, process economics, operations, and scheduling.

Industrial Automation

PLCs, SCADA, OPC-UA, and Modbus, plus the Industrial IoT and data infrastructure that connect equipment to people and algorithms.

Joel kneeling in the front row with the research team in a McMaster hallway, beside a Sartorius research poster
With the research team at McMaster, next to a poster from our Sartorius collaboration. I’m front row, far left.
The research team posed on a glass staircase, with Joel seated in the front row
The team at McMaster. I’m seated front left.

Projects

Selected work

Continuous mAb biomanufacturing · McMaster × Sartorius

In-House SCADA for Continuous Chromatography

I built an industrial SCADA system from the ground up to run the downstream train of a continuous monoclonal antibody process. It connects PLCs over OPC-UA and Modbus and uses Redis and MongoDB as a real-time data backbone, with automated alerts when the process drifts.

  • Controls 250+ valves and streams 8,000+ process variables every 2 seconds
  • Served as DSP lead, deploying and operating the platform during an integrated continuous campaign: 2 L perfusion upstream feeding Protein A chromatography with inline UV/Raman
  • Wrote SOPs and technical documentation for the experimental, automation, and spectroscopy workflows
PLCOPC-UAModbusRedisMongoDBAutomated alerts

Data infrastructure

OPC-UA Server for Raw Spectroscopic Data

I engineered a Dockerized OPC-UA/UDP pipeline that expanded wavelength acquisition from 10 to 201 channels (about 20×). That made full-spectrum PCA/PLS modelling possible on what had been a heavily down-sampled signal.

DockerOPC-UAUDPPCA/PLS

Estimation & modelling

State Estimation of a Nonlinear Chromatography System

I calibrated a mechanistic, ODE-based chromatography model to experimental data and wrote it as a dynamic state-space model. Then I implemented custom Extended Kalman and particle filters to estimate bound and free protein concentrations in real time.

EKFParticle filterODE modelsState-space

Industry · Nylene Canada

Real-Time Wastewater BOD Estimation

I implemented real-time estimation of biological oxygen demand (BOD) to monitor and control wastewater released from a nylon-6 plant. Operators got a continuous view of discharge quality, which supported environmental compliance.

Soft sensingEnvironmental compliance

Optimization in the wild

Camp Elective Scheduling with Linear Programming

I formulated and deployed a Pyomo LP that assigns 4 of 20+ electives to each of 800+ campers. It maximizes preferences and keeps friends together while respecting capacity and scheduling constraints.

PyomoLinear programmingScheduling

Experience

Where I’ve worked

  1. 2024 – Present

    Graduate Researcher

    McMaster University · in collaboration with Sartorius Stedim Biotech

    • Develop, integrate, and operate upstream and downstream automation and connectivity systems for continuous monoclonal antibody biomanufacturing
    • DSP lead for an integrated continuous manufacturing campaign
    • Author SOPs and technical documentation, including UV/Raman spectroscopy workflows
    • Student member of the McMaster Advanced Control Consortium (MACC), presenting at technical meetings and conferences
  2. 2025 – 2026

    Boys Camp Co-Director (Volunteer)

    Galilee Bible Camp · Haley Station, ON

    • Co-organized a 6-day overnight camp week for 100+ campers aged 12–15
    • Supervised 23+ cabin group leaders with daily direction and real-time support
    • Resolved behavioural and logistical challenges and coordinated communication across leadership, support, and program staff
    • Built a Pyomo-based LP to assign electives to 800+ campers
  3. 2022 – 2023

    Process Engineering Intern

    Nylene Canada · Arnprior, ON

    • Validated process modifications to a nylon-6 plant that reduced its carbon footprint and raised production, saving ~2–3%
    • Implemented real-time biological oxygen demand (BOD) estimation for wastewater release monitoring and control

Skills & Attributes

What I bring

Process Control & Modelling

Model predictive controlDynamic process modellingState estimation (EKF, PF)PCA / PLSPID tuning

Industrial Automation

PLC programmingSCADAOPC-UAModbusIndustrial IoT

Bioprocessing

Continuous biomanufacturingChromatographyUV / Raman spectroscopyPAT

Programming & Optimization

PythonC#MATLABSimulinkPyomoGAMS / pyGAMSLP · NLP · MIPGit / GitHub

Data & Infrastructure

MongoDBRedisDockerProxmoxTwilio

Professional

Technical writingSOP developmentTechnical presentationsRoot cause analysisQuality assuranceTeam leadership

Education

Academic path

  1. 2024 – Jan 2027

    M.A.Sc., Chemical Engineering

    McMaster University · Supervisor: Dr. P. Mhaskar · cGPA 3.85

    Thesis: Spectroscopic Driven Estimation of Concentration for Continuous Antibody Manufacturing. Expected graduation January 2027.

    Multivariate Statistics for Big Data (A+)Biopharmaceuticals (A+)Optimization of Chemical Processes (A−)
  2. 2019 – 2024

    B.Eng., Chemical Engineering

    McMaster University · Process Systems Engineering · cGPA 3.70 · Dean’s Honour List

    Coursework covered LP, NLP, and MIP for design, economics, and scheduling; dynamic modelling and model predictive control; and PID and computer process control.

    Optimization in Chemical Engineering (A+)Digital Computer Process Control (A−)Process Control (A−)

Beyond the Lab

Interests & hobbies

Running

A regular racer from 10 km to the full marathon. Training is its own optimization problem: steady effort, careful tracking, and small gains that add up over months.

  • Tartan Ottawa International Marathon4:52:46
  • TCS Toronto Waterfront Half Marathon1:57:31
10 kmHalf-marathonMarathon

Music

Music has long been a creative counterpoint to engineering. It rewards disciplined practice and attention to detail, and it’s a reminder that precision and expression go together.

Camp Leadership

Co-directing a week of camp for 100+ teens and supporting a team of 23+ leaders. It taught me as much about communication and staying calm under pressure as any engineering course.

Joel kneeling in a green team shirt with his team on a sunny campus lawn at the Chem Eng Games 2026
Team “The Reject Stream” at the Chem Eng Games 2026. I’m in the green shirt, front row.

Contact

Let’s connect.

Based in Ottawa, Ontario. I’m always happy to talk about process control, automation, continuous biomanufacturing, or running. Reach out.

joelbaarbe@gmail.com (289) 685-9975 LinkedIn