With the advances in technology, clutter of mechanical indicators in the aircraft cockpit is replaced with digital displays. This revolution does not make only visual changes, but also changes the use of the cockpit design. Cockpit automation has changed cockpit design philosophy with many promised benefits such as improvements in the precision, improved system safety, efficiency of operations, less workload etc. However, to achieve perfect design has not been fulfilled yet. Despite providing innovation and easiness, cockpit automation brings about some &ldquo / Human Factors&rdquo / problems because of lack of support of human-machine interaction and cooperation. In this study, advantages and disadvantages of the cockpit automation will be discussed according to a survey that is conducted to pilots who fly with automated cockpits in Turkey about how automation affects them. The main purpose of this study is to contribute to the modifications of current cockpit systems and development of new design philosophy for advanced flight decks by gathering data from pilots&#039 / attitudes on cockpit automation philosophy.

Although, user requirements are critical for the conformance of a system (or a product) design with the user, they may be appraised late in the development processes. Hence, resources and schedules may be planned with the limitations of system oriented requirements. Therefore, late discovered critical feedbacks from the users may not be reflected to the requirements or the design. The focus of this thesis is how to determine the user requirements of first-of-a-kind interactive systems, early in the development process. First-of-a-kind interactive systems differentiate from others for not having experienced users and subject matter experts. Cognitive analysis techniques are investigated with the aim to discover and integrate user requirements early in the development processes of first-of-a-kind systems. Hybrid Cognitive Task Analysis, one of the cognitive analysis techniques, is carried out for the determination of user requirements of a system in the Human Robot Interaction area. Therefore, while exemplifying the methodology, its competency and correspondence with the domain is observed.

Vibration in the vehicle environment has long been investigated considering the driver&#039 / s exposure. However not only drivers but also the front seat car passengers are exposed to considerable amount of vibration. In order to investigate the phenomenon, this research consisted of three stages. In the first stage of the study, model analysis has been carried out. Based on the results it has been suggested that increased damping in the lumbar area in contact with the backrest can decrease the vibration transmission. The second stage comprised of the laboratory studies. Based on the results attained from the model, waist belts filled with different fluids having different coefficients of viscosity were prepared and tested. The inclination of the backrest angle was chosen as the second parameter. The cushions having ready-made gel mediums were seen to be effective in reducing low frequency vibrations where the angle of inclination affected the response of the cushions. In the third part of the thesis, field measurements were carried out in order to confirm the results attained in the laboratory. It was observed that the inclination of backrest angle played a major role in the exposure of the passenger in fore and aft direction. The cushions proved to be effective at certain frequencies in the field, differing due to the design and the medium.

The physical dimensions of a population change in the course of time. On the other hand, the physical dissimilarities among different population are much more significant and remarkable than time-dependent natural evolution process that affects all populations in the world. In this study, passenger seats of intercity coaches that are important industrial products for Turkey in terms of usage frequency and their prevalence were investigated from an ergonomic viewpoint. To achieve this aim, eight seat parts and their equivalent anthropometric variables were specified to gather their measures from four different intercity coaches: These coaches represent the normal and top segment in Turkish intercity coaches market in addition to being imported or not. After that, these seat measures are compared with two anthropometric studies in terms of the sufficient sample size to reflect the anthropometric data related to Turkish population and the inclusion of the specified variables that correspond to these parts. Finally, alternative dimensions were recommended for each seat part to be used in the design process of passenger seats. In conclusion, it was found that dimensions of all seats demonstrated no significant differences among each other in terms of belonging to middle or luxury segment coaches. Only for seat pan widths, all seats had higher values than the hip breadths of both anthropometric studies. Conversely, for other seat parts, the dimensions of almost all seats had lower values. All backrest heights were the lowest among the other parts based on the studies of Hertzberg et al and KayiS. It was observed that although all seats showed approximate values with specified seat parts of the directive of European Commission- 2001/85/EC- generally, all seat brands except Grammer had a lower value than the seat pan width mentioned in this regulation. It was also found important differences between the outcomes of two Turkish studies and the regulation of EC.

In this thesis, we consider the Multi Resource Agent Bottleneck Generalized Assignment Problem. We aim to minimize the maximum load over all agents. We study the Linear Programming (LP) relaxation of the problem. We use the optimal LP relaxation solutions in our Branch and Bound algorithm while defining lower and upper bounds and branching schemes. We find that our Branch and Bound algorithm returns optimal solutions to the problems with up to 60 jobs when the number of agents is 5, and up to 30 jobs when the number of agents is 10, in less than 20 minutes. To find approximate solutions, we define a tabu search algorithm and an &amp / #945 / approximation algorithm. Our computational results have revealed that these procedures can find high quality solutions to large sized instances very quickly.