Case Study: Innovative Systems
Machine Learning in
Manufacturing
Machine Learning in
Manufacturing
Machine Learning in
Manufacturing
Customer: Covestro, Digital Technical Solutions (dts)
Industry: Chemicals, Manufacturing
Customer: Covestro, Digital Technical Solutions (dts)
Industry: Chemicals, Manufacturing
Overview
Overview
Overview
dts was a system that leveraged IoT systems, computer vision along with machine learning (using TensorFlow) to predict the quality of polyurethane in metal panel production.
With many material experts retiring and not being replaced, Covestro needed to develop a system that could efficiently gauge quality and quantity of Covestro materials within metal panel production lines.
Approach
Approach
Approach
With the project already running, the dts team approached me to simplify the technical messaging for the mass market
I took it upon myself to introduce storytelling and to create a narrative that would resonate with the internal decision board (this was an internal venture). Also did a features review, updated pricing model, customer value grid. Updated personas and customer journeys.
Top 5 Responsibilities
Top 5 Responsibilities
Top 5 Responsibilities
Strategic Messaging and Storytelling: Developed a compelling narrative to simplify complex technical details, making the project more accessible to internal stakeholders and aligning it with Covestro’s broader strategic goals.
Brand Development: Rebranded dts from an internal startup concept to a market-ready solution, ensuring it resonated with both technical and non-technical audiences.
Pitch Deck Optimization: Transformed lengthy technical presentations into concise, impactful pitch decks, highlighting key concepts for a wide audience, including internal decision-makers and potential investors.
Pricing Model and Value Proposition: Redesigned the features grid and introduced a 3-tier pricing model, aligning customer value with pricing and making the business case for the dts system more attractive.
Customer Personas and Journey Mapping: Updated customer personas and journey maps to better reflect market needs, ensuring that the system's development aligned with user expectations and the overall value proposition.
Strategic Messaging and Storytelling: Developed a compelling narrative to simplify complex technical details, making the project more accessible to internal stakeholders and aligning it with Covestro’s broader strategic goals.
Brand Development: Rebranded dts from an internal startup concept to a market-ready solution, ensuring it resonated with both technical and non-technical audiences.
Pitch Deck Optimization: Transformed lengthy technical presentations into concise, impactful pitch decks, highlighting key concepts for a wide audience, including internal decision-makers and potential investors.
Pricing Model and Value Proposition: Redesigned the features grid and introduced a 3-tier pricing model, aligning customer value with pricing and making the business case for the dts system more attractive.
Customer Personas and Journey Mapping: Updated customer personas and journey maps to better reflect market needs, ensuring that the system's development aligned with user expectations and the overall value proposition.
The dts team was burdened with so many moving parts to get the project moving, technically, and working with their pilot program and customers, that they had very little time to work on connecting with a narrative that would tell an emotional story about what they wanted to do.
I started to clean up the presentation and building a brand around dts - taking it from an internal startup idea to a ready for market technical powerhouse. The messaging needed to be simplified for a large audience for investment and board buy-in. Up to this point, the main players on the team were focused on the technical implementation, building out the system and working with their pilot customers.
Taking 100 page pitch decks and reducing them down to a few key concepts for a wide audience:
Updated features grid and development of 3-tier pricing model based off of inclusion of features.
Updated features grid and development of 3-tier pricing model based off of inclusion of features.
The Story
The overarching story was essentially the ability to embrace change in a changing world. In other words, "Old School to New School."
The Story
The overarching story was essentially the ability to embrace change in a changing world. In other words, "Old School to New School."
The Story
The overarching story was essentially the ability to embrace change in a changing world. In other words, "Old School to New School."
A key challenge for Covestro was the retirement of many of its material experts, without plans for direct replacements. These experts traditionally visited material processing plants to optimize material composition, leveraging their experience to account for environmental factors like moisture and heat in order to achieve the best results. The goal of the dts system was to replicate this expertise by training machine learning models to perform these assessments at scale.
The narrative I developed not only resonated with the end user but also helped clarify the core value of the system, independent of the many technical components required for the full dts system to function effectively.
Covestro had an opportunity to capture and commercialize this expertise, even without deploying the complete dts system. By training the models, this expertise could have been offered as a standalone module or as an add-on for integration with third-party providers like Oden or Relimetrics.
A key challenge for Covestro was the retirement of many of its material experts, without plans for direct replacements. These experts traditionally visited material processing plants to optimize material composition, leveraging their experience to account for environmental factors like moisture and heat in order to achieve the best results. The goal of the dts system was to replicate this expertise by training machine learning models to perform these assessments at scale.
The narrative I developed not only resonated with the end user but also helped clarify the core value of the system, independent of the many technical components required for the full dts system to function effectively.
Covestro had an opportunity to capture and commercialize this expertise, even without deploying the complete dts system. By training the models, this expertise could have been offered as a standalone module or as an add-on for integration with third-party providers like Oden or Relimetrics.
A key challenge for Covestro was the retirement of many of its material experts, without plans for direct replacements. These experts traditionally visited material processing plants to optimize material composition, leveraging their experience to account for environmental factors like moisture and heat in order to achieve the best results. The goal of the dts system was to replicate this expertise by training machine learning models to perform these assessments at scale.
The narrative I developed not only resonated with the end user but also helped clarify the core value of the system, independent of the many technical components required for the full dts system to function effectively.
Covestro had an opportunity to capture and commercialize this expertise, even without deploying the complete dts system. By training the models, this expertise could have been offered as a standalone module or as an add-on for integration with third-party providers like Oden or Relimetrics.
