Bootstrap-Based Hypothesis Testing

In this thesis, we explore the structure of consistent bootstrap statistics in hypothesis testing. Bootstrap, as a very useful technique when theoretical distributions are not available or when the sample size is small, enjoys a lot of interest from applied statisticians. Historically, guidelines for performing Bootstrap have been proposed. One...

Methodology for the prediction of the strength of naturally aged glass based on surface flaw characterization

Flat glass is a material which is used extensively in almost all the buildings, mainly as infill for windows or facade panels. However, neither glass production nor its recycling are sustainable processes, as both incorporate the use of gas furnaces, which produce high amounts of CO2 emissions. Thus, the re-use of glass seems the only way to...

Energy Study of Drying: Using Machine Learning to Predict the Energy Consumption of an Industrial Powder Drying Process

In this thesis, we use data science / statistical techniques to better understand the energy consumption behind a powder drying facility located in Zwolle, as part of Abbott's initiative to better manage its energy consumption. As powder drying is by far the facility's most energy intensive process, this project therefore focuses exclusively on...

Unsupervised classification of ships and their operations

In this modern age, data is being generated constantly and data is being saved for analysis everywhere. In the maritime industry, interest in the analysis of ship data has grown over the years. In this thesis, we will take a look at AIS data coupled with sea state data. AIS data is data generated from the ship, concerning the ship locations,...

Probabilistic record linkage with the Fellegi and Sunter framework: Using probabilistic record linkage to link privacy preserved police and hospital road accident records

Record linkage is the procedure of bringing together information from two or more records that are believed to belong to the same entity. The linking of a pair of records without identifier should be based on attributes both records have in common. The probabilistic framework described by Fellegi and Sunter (1969) can be used for record linkage....

Nearshore currents and swimmer safety in the Netherlands

Artificial beach and foreshore nourishments, as executed widely along the Dutch coast, change the nearshore morphology and current patterns significantly. Over the last decades the size of the nourishments has increased from 0.7 x 106 m3/km to over 6 x 106 m3/km. Groynes which were present along the southern Holland coast since the 19th century...

Conditional simulation for characterising the spatial variability of sand state

Properties of soils are spatially variable and to describe the behaviour of soils as a response to loading, this variability appears crucial in giving the correct range of possible solutions for structure response. Because site investigation techniques only provide exact information at a limited part of the site, random field simulations are...

Modelling the 1775 storm surge deposits at the Heemskerk dunes

After a storm surge in November 2007, older storm surge deposits were discovered in the eroded dunes near Heemskerk, the Netherlands. These deposits undulate in height with a maximum elevation of over 6 m above mean sea level. Luminescence dating suggests that the layers were deposited by either the 1775 or the 1776 storm surge. The aim of this...

A probabilistic model for oil spill volume in tanker collisions and groundings

A probabilistic model for calculating oil spill volume in sea tanker accidents is presented. Accidents are specified as either collisions or groundings. Based on existing accident data, damage size is estimated given an accident. Using damage size a probability of rupture is calculated. Finally, outflow volume is determined.