Crimes involving weapons have increased in the Netherlands by 24 per cent in the past ten years, while the decrease in reported crimes has stabilised since 2018. Given that crimes with weapons are often complex and are often investigated in a multidisciplinary manner, these trends increase the demands for complex investigations within the Dutch justice system and, therefore, on the Netherlands Forensic Institute (NFI) as well. Here, the department of Human Biological Traces (BiS) examines evidentiary pieces ("Stukken Van Overtuiging', SVOs), such as knives, to find human biological traces, a process that is called the preliminary examination. The preliminary examination is labour-intensive, and with the increasing demands, could benefit from an optimised workflow. However, to identify and improve issues with the workflow, the workflow needs to be known.
In this observational study, all actions during the preliminary examination were observed and measured. These data were used to determine the duration of SVO examinations and laboratory sessions, their parts and their occurrences. The median duration of a laboratory session was 69 minutes, and the median duration of a completed SVO examination was 29 minutes. The duration of an SVO examination is affected by the category of the SVO, the difficulty level of the examination and the research question. Issues of the workflow were categorised as solvable in the short and long term. In the short term, the workflow could be improved by better managing materials and electronic devices. Furthermore, Processing sample kits could be less labour-intensive by adding codes to the sample containers and performing the examination on a work surface with ridges. The availability and reachability of experts may be improved by aligning their availability with the examination. By solving the bottlenecks concerning material management and processing sample kits, respectively, 15.9 and 56.7 hours per year could be won based on the examinations performed in 2024. In the long term, administrative tasks and examinations with the Crime-lite need further research to provide insights to save time.

Every year, 14 to 41 cases per 100,000 infants under 1 year old are diagnosed with inflicted head injury (IHI), primarily resulting from shaking trauma (IHI-ST) or blunt force trauma. Without reliable structural bending stiffness data of the infant’s neck, injury predictions using physical infant surrogates in shaking simulations remain highly uncertain, undermining both forensic and preventive studies. Due to ethical constraints, biomechanical properties of infant necks are scarce, limiting the biofidelity and validation possibilities for existing infant surrogates, such as anthropomorphic test dummies (ATDs). Currently, infant neck surrogates suffer from inadequate biofidelity concerning stiffness and validated range of motion, necessary to accurately simulate shaking trauma simulations.
This thesis aims to address the gap by exploring the design of a durable, adjustable stiffness surrogate neck, improving the accuracy of shaking experiments. Experimental stiffness values obtained from functional spine units (FSU) by Luck et al., extrapolated by Sullivan et al. were used as target values, suggesting stiffness ranges of 0.2 Nm/rad in flexion and 0.4 Nm/rad in extension for a 1.5-month-old infant. The design aims for a 90-degree ROM in flexion and extension, essential for accurate simulation of chin-to-chest and occiput-to-back contacts, both critical for assessing injury mechanisms.
A compliant monolithic hinge mechanism from Fowler et al. was proposed as the core mechanism, able to achieve large angular displacements through flexures. Finite element analysis was performed to optimize material and geometric parameters. Parametric modeling in ABAQUS identified the relationships between stiffness, stress, material properties, and hinge geometry. Based on these relationships, feasible prototype geometries were extracted, varying in flexure thickness, length, and width. Manufacturing was done using 3D printing with polylactic acid (PLA) and carbon fiber-reinforced polyethylene terephthalate glycol (PETG-CF).
Static experimental validation demonstrated achievable stiffness values at the lower boundary of target ranges while the upper bound was not reached, primarily due to anisotropy from 3D printing and material limitations. Despite these limitations, prototypes successfully reached the targeted ROM. Future research should incorporate dynamic testing to validate durability and head kinematics and should consider multi-degree-of-freedom designs to fully replicate infant neck biomechanics. Further challenges remain in replicating infant neck viscoelasticity and obtaining experimental infant data for validation.

The increasing number of explosive-related incidents highlights the need for objective techniques to characterize and identify pyrotechnic materials, including explosives. Current tech- niques, such as the "hot needle test", rely on subjective human ob- servations of flame color, combustion duration, and intensity, which limits reproducibility and accuracy of the data. In this study, the previously developed Pyrotechnic and Explosive Materials Analysis Device (PEMAD) and its analytical workflow were improved, cali- brated, and validated. Combustions of compositions with varying grain sizes, oxidizer/fuel ratios, color compositions, and sample volumes were recorded under controlled conditions. The camera was calibrated against reference values from a spectrometer and a light-meter. Color calibration reduced the deviations between the camera and the spectrometer measurements from 27-35% to 2- 3%, enabling accurate flame color characterization. Intensity cali- bration allowed pixel values to be expressed in Lux, providing inter- pretable and comparable results across all fibers, although with lim- itations. Validation confirmed that the PEMAD detects differences in combustion duration, flame color, and intensity for compositions with varying properties. Detection limits were established: slower combustions, such as gunpowder, could be reliably measured, while extremely fast combustions, such as flash powder, exceeded the upper limit for accurate peak intensity estimation. The observed sensitivity to small variations in sample volume underscores the need for strict and consistent sample preparation. With larger datasets, the PEMAD could serve as an objective method for the identification and classification of unidentified explosive materials in forensic applications

Inflicted Head Injury by Shaking Trauma (IHI-ST) is a form of abusive trauma caused by violent shaking, which children under the age of one are particularly vulnerable to. Because the exact injury mechanisms of IHI-ST remain unclear and direct evidence such as eyewitnesses is often lacking, legal cases concerning suspected IHI-ST often debate the actual cause of the diagnosed head trauma: Was it shaking, or was it an alternative “accidental” scenario raised by the defendant? One scenario that is sometimes raised is that of a rough trailer ride over bumpy terrain, where the transported infant was exposed to repeated shaking for the duration of minutes to hours. However, the validity of this alternative scenario has not been studied, yet. Therefore, the aim of this study was to evaluate the plausibility of the trailer scenario as a cause for IHI-ST. To get insight in the plausibility of this scenario without performing ethically questionable experiments, we recorded the complete head and torso kinematics of an infant dummy during controlled bicycle trailer rides over bumps with different heights, slopes, and at different trailer speeds. Due to limited biofidelity of the surrogate and the absence of valid thresholds it is currently impossible to know exactly which linear and angular accelerations and which degree of rotational nature of head motion would lead to injury. Therefore, we compared the recorded kinematics of the trailer scenario with recently reported values in the same instrumented dummy under violent shaking circumstances, assuming that the shaking scenario leads to injury. We found that the head motion during rough trailer rides in worst-case conditions (minimized impact damping and the absence of a head rest) was, even at the highest speeds and steepest and largest bumps, less rotational than violent shaking, and while the torso reached similar peak linear accelerations, the head angular accelerations barely reached half of those during violent shaking. However, although the head angular accelerations were lower, a low injury risk for head and neck injuries due to trailer rides can not be concluded yet, because the duration of the infant’s exposure to these accelerations is much longer (minutes to hours) than during violent shaking (a few seconds).