Our MOMS team was transferred in 2019 into a new wholly owned subsidiary, Minopex Technical Advisory. Since then we have been continuing with our MOMS solution, which incorporates data from various data silos into a single hierarchical model of operations and provides BI reports and live displays to stakeholders. We are in the process of migrating our BI reports from Excel to Power BI. A major report consolidation process will follow. Our report developers will coordinate among themselves to deliver report content to various stakeholders in the business without unnecessary overlap or contradictions. Report consolidation, report design, data design and data integration are key focus areas. We route data from on premise sources via Power BI on premise gateway to the Power BI service in Azure. I am closely involved in the data flow design and data integration technologies, as well as forming policy for structured storage of Power BI report definitions files and published content. I occasionally assist in the development of Power BI reportsThe system is constantly evolving. For example, we are evaluating alternative technology stacks. We are also developing our own unified web service-based portal for access via a web browser to all relevant MOM/MES reports, dashboards, and real-time displays.

I am the architect and senior systems engineer of MOMS (DRA Mining Operations Management System). This operations management system is guided by ISA-95, and is a manufacturing operations management system designed specifically for mining & minerals processing operations at ISA-95 Enterprise Level 3. MOMS integrates PI System, PI Vision, and Microsoft SSAS into a system of systems, which interfaces with various external data sources at Enterprise Levels 2, 3, and 4. We have developed our own PI SQL layer with our own custom SQL data warehouse instead of using OSIsoft's PI Integrator for Business, because of our specific custom requirements. Microsoft's Analysis Services on premise supports our BI reports, and hosts the Minopex site business semantic cube models. Our solution follows our own solution pattern, which is repeatable from site to site, while being flexible enough to accommodate site specific requirement. I have pioneered this PI SQL layer and continue to support it on the various deployments.We have developed our own automation scripts in PowerShell, targeting the AFSDK of PI System. These scripts assist with some PI Analytic activities. Further PowerShell scripts monitor basic OS parameters and raise alerts in case of trouble. I have designed and implemented all these PowerShell tools and remain the custodian thereof. MOMS has gone a long way since I redesigned it during the second semester of 2016. We have deployed at Zimplats, covering 4 sites, including mining operations, concentrators, and a smelter. For this deployment we have developed an innovative mining operations model in PI System. We are furthermore deploying at Letseng GEM Diamonds and Phola Coal, where the respective projects are in various phases of deployment. We have evolved our own Agile approach to our solution roll-out and feature development, using Jira Project to track our tasks. I play a leading role in the policies and administration of our Jira content.

I am responsible for the computational capability at the Process Engineering department. In this role, I administrate the computational server stack and NAS server of the Department. I manage the MATLAB & Simulink license pool and distribution, distribution of Aspen Plus, research data management (backup, archiving, etc.), and SharePoint site. I collaborate with the University's IT Unit and coordinate our departmental efforts on computational capability with the University's central ICT plan for research support. We have deployed our new Synology RS2418RP+ NAS server, which is in the midst of commissioning. Our little server farm, hosted on VMWare and managed with vCenter, is doing a splendid job at facilitating research at our department. The Windows Server instances with Terminal Services allow researchers to utilise additional computing power, or reach for a bigger hammer without having to purchasing expensive individual desktop machines. We maintain high availability (ESKOM load shedding permitting) over the course of the academic year.

Chief Engineer at Centre for Process Monitoring, Stellenbosch University:Project co-ordinator:I co-ordinate projects at the Centre; interact with clients to negotiate requirements and milestones; plan and report on yearly programme.Senior software developer: I lead and oversee the development of process performance monitoring tools and dashboards, written mostly in MATLAB; integration with .NET applications and data historians, such as PI System. These tools implement research output of the Process Monitoring and Systems research group.I was responsible for the redesign and development of a MATLAB framework for operational performance monitoring at Anglo American Platinum. I managed the project, did system design as well as implemented the solution. I developed a MATLAB interface to OSIsoft PI AF, targetting the PI AF .NET SDK. The framework connects to process historians and performs predefined data analyses and reporting on operational processes in precious group metal (PGM) plant hierarchies. It plays a key role in monitoring production of platinum at Anglo American Platinum. I provide guidance to colleagues on project matters, as well as programming and engineering issues; assistance to post-graduate students on computational and optimisation problems.Administration of computational and data storage infrastructure:I have established and administrate a virtual farm of servers at the Centre, in co-operation with the on-site IT service unit. The servers provide computational, data management and storage services to the Centre and the Process Engineering Department. I provide technical support to particular researchers at our department in terms of computational capability.I have been involved in the development of a MATLAB distributed computing cluster in Microsoft Azure cloud as well as in the local HPC cluster on campus. In addition, I explore and provide Azure VM solutions to particular researchers at the Engineering Faculty.

Software developer for Nexxim circuit simulator:I co-developed with Deepak Ramaswamy, and maintained, a set of C++ interfaces to the Netlib LAPACK library for solving and manipulating dense matrices in the Nexxim simulator.I assisted Michael Tsuk, a senior scientist at Ansoft, in the implementation of key S-parameter technology for Nexxim.Upon request by the software manager, I designed and developed a core tool set, which formed part of a test framework for the qualification testing of Nexxim on integration level.Configuration manager for Nexxim:I managed and maintained the version control system for Nexxim; provided application support to developers in all aspects of version control. I was responsible for branching and packaging of Nexxim before final integration during each release cycle. I pro-actively proposed and implemented improvements to the configuration management and packaging of Nexxim in co-operation with the Ansoft Designer Team.Build Master for Nexxim:I acted as build master for Nexxim across five platforms, including Windows IA32, EM64T, flavours of Linux and Solaris. Leveraging the superb capabilities of the development team, he led and participated in the porting of Nexxim to build under Solaris 32bit and 64bit as well as Windows XP64 operating systems.

I joined the MATLAB Language Team to participate in developing the core libraries of the M language. During my employment, I proposed and established a software change control policy and mechanism for the Language Team. I overhauled the XLS file interface; created a highly flexible TXT file interface, suitable for importing large text files, internationalized the date and time tools, and extended the file system tools.

I developed a MATLAB toolbox called, QuickIdent, that captured the methodology developed during my doctoral studies.