[en] High quality radiation dosimetry is for workers who rely upon personal dosimeters to record the amount of radiation to which they are exposed. Radiation physicists have been exploring thermoluminescence dosimeter (TLD) for personal monitoring since the mid 1960s, although, widespread use has only occurred in the last 20 years as automated analytical systems and high quality TLD crystals became commercially available. nowadays, multiple TLD (thermoluminescence dosimeter) chips with appropriate physical filters are generally used for measurements of the personal dose equivalent quantities, H_p(d). Though the TLD offers several advantages not possessed by radiological film, it does not offer the some type of advantages as films: re-analysis of an exposure situation is prohibited because the analysis process clears all of the useful dosimetric traps and a record of the luminescence intensity in the form of a glow curve is all that is available after analysis. In addition, the high heating temperatures restrict packaging methods and prevent competitively priced thin films of TLD crystal powders. Optically stimulated luminescence (OSL) technology avoids many engineering limitations imposed by the high heating temperatures used for TLD technology. OSL crystalline powders can be dispersed in various plastics unable to withstand the TLD heating regimen. With uniform dispersion in the plastic, mass-manufacturing techniques can produce large quantities of identically performing detectors. The first proposal conducted by Markey et al. for applications and potentials of α-AI₂O₃:C for OSL dosimetry opened a new era for this phosphor. Pulsed and continuous wave OSL studies carried out on α-AI₂O₃:C have shown that the material seems to be the most promising for routine application of OSL for dosimetric purposes. The main objective of this study is to develop a multi-area personal OSL dosimetry system using α-AI₂O₃:C by taking advantage of its optical properties and energy dependencies. This was done by designing a multi-element filter system for powder layered α-AI₂O₃:C material and an optical reader system based on ultra bright blue LEDs. The main feature of the proposed OSL dosimetry system is that with an appropriate pulsed stimulating scheme and dose assessment algorithm, the personal dose equivalents, H_p(d) can be determined more efficiently and precisely. This dissertation includes various numerical and experimental methods used to design and optimize the performance of the proposed OSL dosimeter may be unfolded from a collection of OSL light emissions following a sequence of optical scanning and dose assessment algorithm. Since the main objective of this work is to obtain the optimum dosimeter system that allows successful measurement of deposited energy distribution, the element's response given as a function of incident photon energy was simulated using a particle transport model, which is calculated using electron/photon Monte Carlo code, MCNP4A. The filtered element responses thus obtained were then used together with angular dependences to design a prototype of the OSL dosimeter. Finally, the experimental response of the designed OSL dosimeter is compared with the original exposure, with good agreement indicating an appropriate dosimetry scheme. Based on the experimental response test of the proposed dosimeter design, it was demonstrated that a multi-area dosimeter system with an LED technology based on α-AI₂O₃:C is suitable to obtain personal dose equivalent information on the mixed radiation fields. With the experimental conditions, the minimum measurable dose was obtained to be 0.1 mGy and that is smaller than the values reported previously. Furthermore, The pulsed blue-LED reader system seems to be quite convenient for OSL measurements from α-AI₂O₃:C and the luminescent output per absorbed dose is larger than the green-LED based system. Therefore, the OSL dosimetry system doveloped in this study can be considered as a compact, reliable and inexpensive system for personal dosimetry. Another work scope of this study is to propose a new dose assessment algorithm for developed multi area OSL dosimetry system. The use of multi-clement dosimeters is necessarily required of a dose assessment algorithm that weights the reading of each element for the evaluation of the personal dose equivalent. In this study, a feed forward neural network using the error back-propagation method was applied for the response unfolding procedure. To replace the simplistic decision tree algorithms by the more sophisticated neural networks, the Bayesian algorithm was introduced to optimize the MLBP (Multi-Layer Back Propagation) neural network. The validation of the proposed algorithm was investigated by unfolding the measured responses of the OSL dosiemter for arbitrary mixed photon field range from 20 keV to 662 keV. Results show that the determined dose values are reproduced with an accuracy better than 89%. With these validation results, it was demonstrated that a multi-element dosimeter system with an unfolding algorithm based on artificial neural intelligence is suitable to obtain spectral information on the incident gamma photons. It was also found that the Bayesian's approach was helpful to optimize the connective weights to solve the abnormal function overfitting. For the unfolding of the multi-element response, merely these weights were applied without carrying out further optimizations. In consequence, the unfolding step requires less than one second, i.e. several orders of magnitude less of computational time in comparison to the optimization step. Therefore, the response unfolding can be carried out frequently and can be implemented either as hardware or as software into the measuring instrument in this topology. The novel approach presented in this study enables the development of the next generation personal dosimetry system using a pulsed optically stimulated luminescence technology