Water management infrastructures such as floodgates are critical and increasingly operated by Industrial Control Systems (ICS). These systems are becoming more connected to the internet, either directly or through the corporate networks. This makes them vulnerable to cyber-attacks. Abnormal behaviour in floodgates operated by ICS could be caused by both (intentional) attacks and (accidental) technical failures. When operators notice abnormal behaviour, they should be able to distinguish between those two causes to take appropriate measures, because for example replacing a sensor in case of intentional incorrect sensor measurements would be ineffective and would not block corresponding the attack vector.

In this thesis, we developed the attack-failure distinguisher framework for constructing Bayesian Network (BN) models which enable operators to distinguish between those two causes, including the knowledge elicitation method to construct the directed acyclic graph and conditional probability tables of BN models.

As a full case study of the attack-failure distinguisher framework, we constructed a BN model to distinguish between attacks and technical failures for the problem of incorrect sensor measurements in floodgates, addressing the problem of floodgate operators. We utilised experts who associate themselves with the safety and/or security community to construct the BN model and validate the qualitative part of constructed BN model. The constructed BN model is usable in water management infrastructures to distinguish between attacks and technical failures in case of incorrect sensor measurements. This could help to decide on appropriate response strategies and avoid further complications in case of incorrect sensor measurements.@en

In recent years, the Netherlands Ministry of Defence (MoD) modelled, standardized, and digitally secured many of its organization-broad business processes. However, MoD experiences great difficulty in gaining insights in these processes and steering them productively. The novel business process analyses method Process Mining appeared to be a promising instrument. However, MoD lacks experience with process mining as a way to analyse and subsequently improve processes. In addition on how to apply process mining, MoD wants to learn the requirements for starting a process mining project.

The thesis starts with a literature study on eight business process improvement methodologies (BPIs), viz. Lean, Six Sigma, Lean Six Sigma, Total Quality Management, Lean MRO, Total Productive Maintenance, Business Process Reengineering, and Business Process Improvement. These BPIs are compared with each other in an overview based on nine characteristics. Then, a ninth BPI is studied: process mining. The literature on this topic is scattered, but a study is conducted on several element. After that, pros and cons are discussed and process mining is placed in the overview of BPIs.

A combination of desk and field research leads to the improvement of the best practise methodology PM². And this improved methodology, called PM³, is further within MoD. Next to that, a process mining decision framework is developed according to the waterfall model. This framework can score business process models on several characteristics. The resulting total score determines whether process mining can successfully be applied to improve the process.

For MoD, process mining can be a valuable instrument for improving processes. It can convert a gut feeling into clear numbers and models, which can form the start of an improvement project. It is thus recommended to make process mining available to the organisation.

The thesis gives a good insight in the possibilities of process mining, but it also comes with several opportunities for future research. For example, PM³ has demonstrated itself in the field, but this have not yet led to measurable improvements (only important insights in the processes). Future research is required to verify its true potential. Second, the decision framework is tested by assessing fourteen of MoD’s processes and evaluated by discussion it with ten system experts. Yet again, empirical research is probably the only way to truly verify it, but this was not possible during the thesis. Third, both PM³ and the decision framework focus on MRO within MoD. It is unlikely that a change of domain or organisation makes a significant difference, but the possibility should be taken into account and further investigated. And finally, a significant limitation of the framework is that it only scores on process mining. So, if it presents a very low score, it does not suggest an alternative BPI. This fell outside the scope of the thesis, but is nevertheless an interesting research opportunity.