Synopsis - Inquisitive, Collaborative, Innovative

---

An engineer and scientist focused on leading the development of solutions to deliver innovative and differentiating award-winning products. Striving to solve technical challenges by applying multidisciplinary knowledge to facilitate, lead teams and mentor individuals in their areas of expertise. Passionate about communication, distilling complex ideas into simple messages. Data driven, basing decisions and strategies on a foundation of sound information.

Some of the fun things I've done during my career

---

Data, modelling and understanding the problem

As an engineer and scientist the acquisition, interrogation and communication of data has been an essential part of my successful career. I have primarily worked in and consulted for companies in the pharmaceutical and medical device sectors solving technical challenges.

Examples of my work include:
Leading multiple product verification studies and being responsible for the application of statistical techniques to meet industrial standards. Within the field of IVD’s, I have guided studies meeting compliance to the CLSI standards resulting in successful FDA submissions. I have also supported clients in the successful verification of devices like in-halers and otoscopes.

Integrating experimental data and math models to predict future results and optimise performance has been an essential part of understanding the mechanics of devices during their development. For example:

• Integrating numerical methods to estimate fluidic mechanical systems and exploring scenarios with Monte Carlo simulations to investigate pharmaceutical equipment. This was used to refine the requirements of a system while showing the bottle-necks of the manufacturing process so work packages could be focused on addressing critical areas.
• Providing guidance on statistical evaluation methods and DoE. This supporting colleague's evaluated their scientific hypotheses developing processes for manufacture using standard practices and sound techniques.

Developing and validating physical models and rigs has always been an active part of my product development. For example I've simulated the chemical degradation and mass transport in a drug product's primary packaging to support and guide design decisions enhancing shelflife. Investigations highlighted the primary parameters that needed to be controlled as part of the design and influenced plausible concept solutions.

Other models created as part of design concept, feasibility and optimisation include:

• Modelling electromagnetic fields to optimise magnet shield design for IVD product, resulted in geometry that minimised magnetic interference.
• Simulation of optical devices for root cause analysis of production issues with otoscope light guide. Resulting in the successful identification of the design change that caused the issue so it could be amended.
• FEA stress-strain analysis on concept devices used to optimise designs and influence material selection.

---

Developing devices from concept to market

I’ve led and managed multiple projects through all phases of development from concept through to launch resulting in successful products. Many of these products have been for the pharmaceutical and medical device industries with a strong focus on injection moulding and sheet metal manufacturing with integrated electronics and software. Example product areas include: IVDs, drug delivery, optical instruments and bio-processing equipment.

While developing various products I've successfully created test systems and rigs to verify and investigate a product’s performance. These test systems focused on generating quantitative data for example:

• Creating a performance evaluation rig for disposable IVD cartridges which included fluidic control generating real-time force and distance data with image analysis of fluidic movement. I successfully wrote the rig control and data acquisition software in Python. The rig identified key performance characteristics and batch variations that needed further control in the production process.
• Wrote an image analysis program to fit young-laplace curves to droplets in order to measure contact angles on chemically modified surfaces. This provided a quantitative analysis of surface changes that could be correlated to device performance.

---

Line-management, project management, leadership and mentoring

I have line-managed and been responsible for an engineering department (health & safety, budgeting, forecasting, recruitment) while also leading the technical development. I’ve successfully moulded engineers and scientists into high performing teams with a focus on trust, independence, accountability and supporting each other. I’m also experienced in project management having led, implemented and worked within various implementations of the agile principles and phase-gate development processes while taking products from concept through to market.

Qualified as an engineer (1st class, University of Glasgow, UK) with a PhD in sensor technology (Cranfield University, UK).

Disclaimers and credit

• This website was coded by me (no AI was harmed in the creation of this site)
• All images were taken or made by me (or will be when I finish the site)
• Design ideas, input and direction from the awesome Ted Low (Ted's site)

Report Github

This investigation uses thermal modelling to simulate the Ajitsuke Tamago cooking procedure as the egg changes its stages as it solidifies.

Overview

---

Ajitsuke Tamago, aka Ramen eggs, are one of my favorite additions to a hearty ramen bowl. The general recipe for Ajitsuke Tamago requires adding an egg to boiling water for 7 minutes and then submerging it in ice cooled water to prevent further cooking.

For this investigation a thermal model of the egg was created using finite-differences in python and Jupyter. The model allowed custom meshes which defined the egg geometry and varying diffusion coefficients and also dynamically changed the boundary conditions based on defined criteria.

The results showed an alignment between the generally cooking method and the hardness of the composite egg parts (white and yolk) based on the temperatures reached.

Find out more...

---

Read the full Report. Or review the code in full at Github.

Report Github

Github

Lego is awesome! So when I discovered there was a data available to construct a database based on all the sets made since 1949, I could not resist having a look. So, the first step to exploring this data was to build a local database that I could query.

Overview

---

Construction of the database was done in 5 stages:

• Install MySQL
• Wrote SQL scripts to create DB and build Schema based on image above
• Download and wrangling of CSV files from rebrickable
• Loaded CSV data into DB with SQL scripts
• Ran checks to ensure all was working

Fundamentally I did this because I like Lego! But it also provided me with a local DB to explore, test out my analytical skills and indulge my curiosity. It was a great reminded that cleaning up the data is nearly always required and a significant chunk of any analytical analysis. For example, I assumed (hoped) the data would be in good shape coming from a working system and in the most part it was. However, the data did not match the schema (v3) any MySQL does not like empty cells nor carriage returns which had to be addressed.

Find out more...

---

Review the notes and access the scripts used on Github.

Github

Report Github

This investigation looks into a Lego brick database to explore how lego bricks have changes since it conception in 1949.

Overview

---

Lego is an award winning product and has been popular for decades. Out of intrigue I create a MySQL database from rebrickable datasets. This investigation is my exploration into this database.

This investigation primarily uses R and SQL to look at the 273 different colours of bricks, trends identified in the early 2000's when the company announced it's 1.9 billion kr deficit and other observations including Fabuland Lime toilets.

Find out more...

---

Read the full Report. Or review the code in full at Github.

Report Github

Highly toxic drugs are often used to combat tumours, however the high potency can be detrimental to healthy tissue. Understanding how the drug compounds diffuse through the tumour can help to reduce the chances of damaging healthy tissue.

Overview

---

This investigation looks at the image analysis of tumors to identify their constituent parts. Then the transport of drug product can be simulated assuming a known point of injection.

The springboks have dominated international rugby since the 2019 world cup. This is exploration to look for any insights or correlations that may indicate why this is.

Overview

---

This investigation consists of:

• Acquisition of data from various websites
• Data wrangling resulting in a ordered data frame(s)
• Initial exploration of the data to get a feel for what is there
• Use of an unsupervised ML algorithm to identify any clusters in the data
• Look to see if any parameters can be correlated to success

Interactive chart Report Github

This is a exploration into using the World Health Organisation (WHO) OAPI to access their Global Health Observatory (GHO) database.

Overview

---

The Global Health Observatory is the World Health Organisation's gateway to share data on global health. It is structured into categories of indicators (3043 at time of writing) that describe health statistics. They generally all have dimensions of time and country, some have additional dimensions for example 'Age Group', 'Sex' and 'Severity'.

This exploration primarily uses python to explore the GHO databases with GeoPandas for generating geospatial charts. Example interactive charts were made using D3 to demonstrate how the WHO data could be visualized.

Find out more...

---

Use an interactive chart, read a Report . Or review the code in full at Github.

Interactive chart Report Github

Report Github

This simulation focuses on modelling pendulum systems with the future intention of implementing a control feedback system. Consequently, a function was written where time-dependent ODE systems can be solved in a step-wise manner.

Overview

---

I've been fascinated with control systems since seeing an inverted pendulum on a cart during a university open day. Although I studied control feedback systems I've never modelled this classic example, so this is the start of my inverted pendulum on a cart simulation.

Modelling pendulum systems stared by writing a time-dependent ODE function in python and Jupyter that can be solved in a step-wise manner. This was based on a solution by Sergey Royz that was modified for future implementation in control feedback systems. A 4th order Runge-Kutta methods was demonstrated using an ideal system.

Find out more...

---

Read the full Report. Or review the code in full at Github.

Report Github

This is looking at control feedback systems for inverted pendulums. In particular chaotic double pendulums.

Overview

---

I've been fascinated with control systems having seen a physical implementation of an inverted pendulum at university. This is my development of a control system for a single and double inverted pendulum using the pendulum simulator that I previously created.

Different machine learning algorithms are selected and then trained for the offensive and defensive sides of a tower defense game.

Overview

---

This is a simulation of a classic tower defense game where an offensive side has to travel from a starting location to the oppositions base, take resource, and then return the resource to the start location. Once all resource is acquired, the game is won for the offensive side. However, the defensive side can build turrets to defend their base giving them the option to win by rendering all the offensive side out-of-the game.