PhDs | yiap

Current PhDs

Connor Walker, Safe LLMs for Maintenance
Razieh Arshadizadeh, Model repair in Executable Digital Dependability Identities
Kuniko Paxton, Machine Learning Fairness in a Continuum

Recently Completed PhDs

Luis Torrao (2024), Timaeus - Tetrahedral Illuminated Media Augmented Scupltures
Athanasios Retouniotis (expected 2023), Model-Connected Safety Cases
Koorosh Aslansefat (expected 2022), Addressing Complexity and Intelligence in Systems Dependability Evaluation
Ioannis Sorokos (2018), Generation of Model-Based Safety Arguments from Automatically Allocated Safety Integrity Levels
Luis Azevedo (2015), Scalable Allocation of Safety Integrity Levels in Automotive Systems
Zhibao Mian (2014), Model Transformation for Multi-objective Architecture Optimisation for Dependable Systems
Shawulu Nggada (2013), Multi-objective System Optimisation with Respect to Availability, Maintainability and Cost
Nidhal Mahmud (2012), Dynamic Model-based Safety Analysis: from State Machines to Temporal Fault Trees
Amer Dheedan (2012), Distributed On-line Safety Monitor Based on Safety Assessment Model
Septavera Sharvia (2011), Integrated Application of Compositional and Behavioural Safety Analysis
David Parker (2010), Multi-objective Optimisation of Safety-Critical Hierarchical Systems
Ian Wolforth (2010), Specification and Use of Component Failure Patterns
Martin Walker (2009), Pandora - A Logic for the Qualitative Analysis of Temporal Fault Trees

Potential New PhD Projects

I am currently accepting applications for PhDs in all areas of my research. For potential PhD projects please see examples below in the areas of HiP-HOPS, Machines Designing Machines, 3D User Interfaces, and Digital & Generative Art. The key requirements for PhD applicants are a good degree, strong software engineering skills, and an exploratory mind. Much of the work has extensive industrial applicability and is done in collaboration with large industrial organisations which are technology leaders in their field. Despite the many applications of this work in the engineering of technologically advanced systems, such as electric and semi-autonomous cars, the work involved in those PhD projects does not require any knowledge of these engineering domains and is mainly focused on computer science, and the development and implementation of novel algorithms and tools using standard computer equipment and programming environments. Please contact me if you are interested in a PhD.

HiP-HOPS - not the music but PhDs in a state-of-the-art method and tool for dependable design

The DEIS research group is presently pioneering the development of novel methods and tools for dependability analysis and optimisation of complex safety critical engineering systems collectively known as Hierarchically Performed Hazard Origin and Propagation Studies (HiP-HOPS). The technique and the tool can help to predict how dependable is a system by automatically analysing design models of a system under development. In the context of a string of recent European projects, HiP-HOPS has also contributed to the specification of EAST-ADL, an emerging architecture description language developed as an automotive industry standard for the design of vehicle control systems. HiP-HOPS today is widely recognised as one of the state-of-the-art techniques in the area of dependability analysis.

Following more than 20 years of research, the HiP-HOPS software tool was commercially launched in 2012. ITI GmbH, a software house and author of the simulation tool SimulationX have integrated HiP-HOPS into SimulationX and provide licences to interested parties. This work has developed with support from the British Government, the European Commission, Volvo Cars, Jaguar-Landrover and many other organisations. For more information about HiP-HOPS tool please visit the HiP-HOPS website.

The project has so far yielded several PhDs and over 100 publications in scientific journals and conferences. This is very much a live scientific development and there are currently more opportunities for doctoral work in this area. These include development of concepts for representation and re-use of component failure logic and patterns, synthesis with emerging Architectural Description Languages for design, concepts for dealing with uncertainty in dependability analysis, and development of algorithms for real-time and on-line fault diagnosis, and health and safety monitoring using intelligent agents

Machines Designing Machines? - How to automatically improve system design models

In a futuristic scenario, machines will be designing new machines.

Is it possible? Is it inevitable? What are the consequences for the economy and what are the ethical implications for society? I am not going to discuss these questions here - perhaps in a PhD supervision instead. :)

However, I would like to focus on the clearly positive aspects of such an automated design capability. Complex computer-based systems such as those used in transport, medical devices and financial systems are becoming increasingly difficult to design in a way that certain important properties related to the dependability of the system can be guaranteed. Reasons include increasing scale and complexity of systems as well as the new difficulties arising from flexible and often error prone implementation technologies.

The aim of this work, which can form the basis of several PhD projects, is precisely to use state-of-the-art technologies that are emerging in the field of artificial intelligence to provide some support in the design of complex computer-based systems by enabling automatic optimisation of initial design models for such systems, e.g. expressed in notations similar to UML, with respect to criteria that include for example, safety, reliability, availability, weight and cost. To achieve this aim, genetic algorithms and other meta-heuristics have been developed to enable the automatic 'evolution' of initial designs drafted by humans to improved equivalents that optimally employ for instance hardware and software replication to improve the dependability of systems with minimal costs.

This work is highly innovative and leads both to theoretical contributions and useful industrial applications. Publications of results so far have received a distinction among best papers in several international scientific conferences. We have published results together with Toyota in the Journal of Software Practice and Experience. And some of our work in this area which was led by Luis Azevedo, one of our PhD students in his second year of the PhD, was featured in the prestigious IEEE Software magazine.

PhD research projects in this area will extend current work on optimisation of system architecture, system maintenance, and optimal allocation of system requirements.

Designing new generation 3D interfaces for system design, analysis and optimisation tools.

In recent years, the DEIS research group has pioneered state-of-the-art tools for design and software engineering of systems such as aircraft and cars. Foremost amongst these is HiP-HOPS, a novel dependability analysis and optimisation tool that helps engineers to develop safer, more dependable systems. HiP-HOPS is a commercially available tool used by major companies like Honda, Volvo, Toyota, and Continental, amongst others. It also has applicability to other critical systems, e.g. those that handle energy infrastructures, telehealth systems, or financial systems.

Our work so far has focused on developing advanced algorithms that exploit advances in artificial intelligence to develop new capabilities in design. However, our experience in this area also suggests that there are great research and commercial opportunities in developing better user interfaces for tools that support these system design processes. The interaction between users and system design tools follows paradigms that were developed in the 80s and are often ill-suited to the expectations of modern users. New modes of interaction supported by haptic devices and interfaces, involving for example large touch screens, could revolutionise the way systems are designed, improving both user experience and productivity.

We have specific ideas on how to enhance collaborative engineering of systems by enabling interaction between users and providing better interfaces to system models and analyses in the course of specific design techniques that are prevalent in many industries. These can form the basis for PhDs that have traction in industrial practice and potential for reachout and commercial ventures. The key requirements for potential PhD applicants are good software engineering skills and an interest in user interface design.

Digital Audio-Visual Art for Aesthetics, Art Therapy, Education and other Socially Useful Applications

We are presently working in collaboration with painters and musicians on a range of exciting new projects in the field of generative and digital art. Recent results include a conceptual artwork which has gained international recognition and was exhibited in La Triennale di Milano, the main contemporary design museum in Italy. This work was also the subject of a paper entitled "An evolving Musical Painting on The Boundary between Permanence and Change" that was presented in the XVI Generative Art Conference.

These projects bring together a number of different disciplines, including software engineering, graphics, and the application of artificial intelligence techniques. It is ideally suited to Hull graduates or any other graduate with an interest in a PhD and some quality training in these fields. The work may include the possibility of developing art installations over networks of interconnected internet-enabled devices (internet of things).

As well as producing and exhibiting artworks with their own intrinsic value, the other primary aim of this innovative research is to investigate how this type of digital art can be used for therapeutic and educational purposes, e.g. in commercial applications or via installation in hospitals and clinics. For instance, feedback by clinicians has suggested that this kind of art can have therapeutic effects on people with long term conditions like autism or dementia.

One of the objectives of this research is therefore the development of digital art platforms that can test this hypothesis and investigate the potential benefits of digital artwork in health-related contexts. We have also planned a series of "musical painting sculptures", including a musical art sphere and a painted Möbius strip, that can be virtualised and enhanced with visualisation technology to create interesting art works that, beyond their artistic merit, can also provide case studies for art therapy. Much of this work can also evolve as educational games with potential reach out in this direction.

There is scope for a number of different PhD projects in this area, so the precise direction of the work is open to discussion and can be adjusted to suit the interests of potential students. There is also the possibility that potential PhD projects will involve colleagues from the Faculty of Health and Social Care, who will help with setting up clinical studies.

VIVALDI (Virtual Audio-Visual Dynamic Musical Instruments)

This project is a component of a larger research theme on generative art. This theme is seeking to produce cutting edge audiovisual interactive digital art, and to investigate applications for art therapy, novel musical instrumentation and educational games. Projects have been presented at international conferences and exhibited in La Triennale di Milano and other venues. Examples can be experienced in this site

The project will explore the creation of virtual musical instruments that can create new and exciting audio-visual experiences and which can expand what musicians can do on stage thus giving a new edge to live performances. The project collaborates with artists and will base these virtual instruments on works of fine art, for instance objects such as painted sculptures specifically created for this purpose. These objects will be virtualised using laser scanners and cameras and will be seeded with sounds to enable the creation of music. Musicians will be able to specify the potential musical design space defined by each instrument, and in real-time manipulate these virtualised instruments using gestures thus creating music and stunning visuals. The 3D immersive technology and laser scanning facilities available in the Hull Immersive Visualization Environment (HIVE) will be used to enable the creation and fine-tuning of these novel musical instruments.

We already develop an exciting project in this area which is inspired by a Platonic dialogue from which it takes the name TIMAEUS standing for ‘T’etrahedral ‘I’lluminated ‘M’edia ‘A’ugm‘e’nted Sc‘u’plture‘s’ Art Studio. TIMAEUS is an open block-world art studio being developed as a system for art therapy for people suffering dementia. The proposed work will be linked to TIMAEUS, and will explore the utility of these technologies within a therapy environment collaborating with the Faculty of Health, Hull Royal Infirmary and Memory Clinics.

We seek candidates with a first class degree in computer science or music technology, programming ability and a good knowledge of music (particularly relating to composition).