Alexander Dade

All programs

Program 2 of 2

Taking a new glass melting technology from pilot plant to production

Case study · O-I · 2019 to 2020 and 2023 to 2024

The melting technology is proprietary. This page covers the program, the methods, and the people, not how the technology works.

At a glance

Role
Program Manager on the pilot plant, 2019 to 2020. Senior Technical Program Manager on the scale-up, 2023 to 2024.
Scope
$250M program · 200+ engineers · 12 portfolio teams · 3 global facilities · 19 countries
Result
Two revenue-producing production plants
My part
Vendor systems integration, the gates, the people plan, safety, and the knowledge the plants run on.

The problem

A first-of-its-kind glass melting technology had to move from experiments to plants that met a quota and turned a profit. Each site was one of a kind.

The production line was assembled from vendor-built automated systems: robotic batch handling with autonomous guided vehicles (AGV), and AI inspection machines that adjusted the line based on what they saw. Both fed one SCADA control layer. My job was to make them run as one line, and to make sure our engineers could run it without the vendors.

I worked on the pilot plant in 2019 and was brought back in 2023 to run the scale-up.

Sequence

  1. R&D labexperiments
  2. Pilot plant2019 to 2020
  3. Scale-up2023 to 2024
  4. Production plantinherited in flight
  5. Production plantto quota

Fig. 2The program from lab to production. Each stage closed with signoffs and demos.

How it ran

Pilot plant to production: the program, in order

  1. Prove it at the pilot plant.

    I stood up the experimental production site, a full-scale glass plant instrumented with hundreds of sensors and connected devices, to prove the technology for scale-up.

  2. Pick up the scale-up.

    I came back to 12 portfolio teams, 200+ engineers, and three facilities. One production plant was already in flight when I took it on.

  3. Gate every stage.

    Every stage closed with signoffs and demos. Construction and commissioning ran Waterfall, anything software ran Agile, and hardware startup and configuration ran hybrid.

  4. Integrate the vendors.

    Dozens of vendors needed onboarding, data access, and system configuration. I qualified each vendor-built system, validated each control loop into SCADA, ran QA validation, and trained engineers to operate the systems and check the vendor’s work. Testing ran on a segregated copy in a test environment, so tests did not touch production.

  5. Plan the people and the assets.

    I forecasted engineering headcount, built the capacity model for every asset, and interviewed and hired for the team. Every asset had a BOM, and cost, depreciation, risk, and ROI were reported through SAP and Microsoft Project.

  6. Run the work in one place.

    A templated Azure DevOps intake saved 8 hours a week per person on a 12-person IT team. A ticketing and service management system cut launch time from a month to under a week. A QA validation standard took project timelines from 25 days to 4 and became the organization’s standard.

  7. Hand over the knowledge.

    I migrated the DITA XML knowledge base to a new platform with Okta SSO for 10,000+ users, live on schedule with adoption up 5x, and rolled out Microsoft Copilot to 300 engineers, cutting onboarding time 35%.

Failure modes and controls

Failure modeControl
Only one person can run a vendor system.I trained engineers on it until qualified coverage went from 1 person to 4.
Arc flash exposure on new electrical equipment.I saw the gap, escalated it to EHS and electrical engineering, co-wrote the protocols and the PPE specification, and trained the team that worked the equipment.
The upstream electrical build runs on its own timeline.I tracked and QA validated it against its schedule and sequenced the process work to start when it launched.
Commissioning shows different parameters than planned.I changed the scope of protocols and processes mid-flight and reworked the runbooks and playbooks to match.
A vendor system behaves differently on the line.Each control loop was validated into SCADA and tested on the segregated copy before it reached production.
Only one person can run a vendor system.
Control
I trained engineers on it until qualified coverage went from 1 person to 4.
Arc flash exposure on new electrical equipment.
Control
I saw the gap, escalated it to EHS and electrical engineering, co-wrote the protocols and the PPE specification, and trained the team that worked the equipment.
The upstream electrical build runs on its own timeline.
Control
I tracked and QA validated it against its schedule and sequenced the process work to start when it launched.
Commissioning shows different parameters than planned.
Control
I changed the scope of protocols and processes mid-flight and reworked the runbooks and playbooks to match.
A vendor system behaves differently on the line.
Control
Each control loop was validated into SCADA and tested on the segregated copy before it reached production.

Results

  • $250Mfirst-of-its-kind program
  • 2production plants, from one pilot
  • 25 → 4days per project, after the QA standard
  • 1 → 4people qualified on a critical vendor system
  • 5xknowledge base adoption

What colleagues said

  • His collaborative approach was essential to expedite the adoption of CCMS and LMS programs that were new for the glass manufacturing environment.

    Javier Reyes, MSc, PMPProject Planning Director · Managed Alexander directly at O-I · December 2024
    Read the full recommendation

    Alex's expertise and strong organizational skills were key to creating technical training materials for innovative technology. His collaborative approach was essential to expedite the adoption of CCMS and LMS programs that were new for the glass manufacturing environment. Alex’s ability to quickly understand complex technologies and ensure their successful rollout makes him a standout professional for any organization focused on innovation and efficiency.

  • He effectively gathered information from various parties and distilled complex topics into easy to understand documents which were used across continents.

    Michael HuTechnology Licensing Lead, O-I · Senior to Alexander · November 2024
    Read the full recommendation

    Alex was instrumental in creating best-in-class documentation for novel manufacturing processes. He effectively gathered information from various parties and distilled complex topics into easy to understand documents which were used across continents. Alex also provided valuable feedback on how to enhance the utility of resources based on his prior experience.