Sharif Digital Repository

This study is about real-time hybrid motion planning for autonomous UAVs in dynamic environments. The algorithm is based on system dynamic quantization to trim trajectories and maneuvers, constructing a library of primitives which guarantee the controllability of the system. Random algorithms introduced in literature of motion planning have an offline phase, reducing the computational complexity of online phase. By using dynamic quantization, we have achieved a new totally online algorithm, increasing the probability of finding a solution. Dynamic programming is the core of this algorithm, instead of offline calculations in before random algorithms. In order to decrease the exponential... 

In this study, the viscous compressible flow is simulated over two-dimensional geometries by using the immersed boundary method and applying a high-order accurate numerical scheme. A fourth-order compact finite-difference scheme is used to accurately discretize the spatial derivative terms of the governing equations and the time integration is performed by the fourth-order Runge–Kutta scheme. To regularize the numerical solution and eliminate spurious modes due to unresolved scales, nonlinearities and inaccuracies in implementing boundary conditions, high-order low-pass compact filters are applied. A uniform Cartesian grid that is not coincident with the body surface is used and the boundary... 

Iran's experience shows that either single or combined power generation cycles, face heavy summer performance deficiencies. The most limitation of power generation in steam cycles is due to their poor condenser performance.The less condenser performance, the more turbine back pressure, and the more turbine back pressure, the less cycle power generation. The main objective of the present study is to reduce steam turbine back pressure in the combined cycle of an organic Rankine-vapor compression refrigeration by installing heat pipes in suitable locations inside the Air cooled condense (ACC). In order to do so, a commercial numerical simulation solver (FLUENT) is used to find the most suitable... 

This thesis introduces and investigates a new kind of scheduling model, in which the total amount of computational resources to be allocated at each moment by the processor is constrained and upper bounded by a constant. The time for a task to be executed is also dependent on the computational resources, dedicated to it by the processor, assuming that the multiplication of the execution time by the dedicated computational resources is a constant for each job. We investigate various objective functions, such as minimizing the number of tardy jobs, minimizing maximum latency and etc, under varying constraints. The investigated objective functions are: a) Minimize number of tardy jobs b)... 

With the growth of the number of online service providers and the need to innovate in these services, in this study, the processes and the journeys taken by visitors of a website that provides employment services and employment insurance has been analyzed. In this research, process mining techniques and predictive process monitoring were implemented. With the use of a supervised and unsupervised learning algorithm, it attempted to identify the customer journeys' output and the existing patterns that lead to the complaint. In the first step, the website event log is extracted. Afterward, by using frequency-based encoding methods, the journeys traveled by users were clustered based on the... 

In order to survive in today's business world, which is changing at a very fast pace, organizations can detect deviations even before they occur, quickly and with a high percentage of confidence, by analyzing their processes, in order to prevent disruptions in the processes. by monitoring the information systems that automatically execute business processes, it is possible to ensure the correct implementation of the existing processes. For this purpose, various techniques for monitoring business processes have been presented so that managers have a comprehensive and real view of how implement processes and be able to identify possible deviations in the future and try to fix them because the... 

In recent years, several studies have been conducted in the field of data mining techniques in the field of process mining with the aim of improving the performance of organizations. Predictive process monitoring is a data-driven approach that helps business managers to improve the status and conditions of their organization. In this approach, the event log, which includes a set of completed examples of a process, is received as input, and machine learning methods are used to predict the outcome and results of the organization's processes before the process is completed. This prediction can include the prediction of the final result, the next event, the time remaining until the completion of... 

Brain Computer Interface (BCI) is a communication system between human brain and a computer or a peripheral device which by recording brain signals directly would send messages and commands from the human brain to computer.According to brain activity patterns of EEG, BCIs are divided into different types. The most important of these patterns called ERP (Event Related Potentials) which appears after particular events in the EEG signal. A significant ERP pattern is P300 potential. It occurs when patient recognizes oddball stimuli. SSVEP (Steady-State Visual Evoked Potential) is another type of patterns and is response of the brain to optical stimulations with certain frequencies and a strong... 

The inventory costs constitute a significant portion of the supply chain costs. Therefore, choosing an optimal inventory policy for orders is of great importance. The aim of this research is to find the optimal inventory policy for a distribution center in a three-tier supply chain consisting of a manufacturer, a distribution center, and a retailer. This research simulates a supply chain in agent-based framework and optimizes it using reinforcement learning. The optimization KPI in this research is the mean daily cost of the supply chain. Finally, the result obtained from reinforcement learning is compared with the optimized result of AnyLogic and the mean daily cost in the model optimized... 

Business processes are an essential part of every business as they provide insights on how to optimize and make them more efficient. Predictive Business Process Monitoring has garnered significant attention in recent years due to its capability to forecast process outcomes and predict the next activity within an ongoing process. In the last few years, there have been works that focused on deep learning and its applications in predicting the next activity. Some research used Long Term Short Memory, while others used Convolutional Neural Networks. However, long term short term memory models have the constraint of relatively slow training, while Convolutional Neural Networks are fast but may... 

The analysis of customer journeys has gained significant attention due to the critical role of customer behavior data in enhancing business decision-making and formulating strategies for customer acquisition and retention. By segmenting customers based on their journey patterns, businesses can offer personalized recommendations, thereby improving customer engagement and loyalty. Additionally, predicting the next steps in a customer’s journey based on historical data allows for timely and appropriate interventions at various touchpoints. By understanding where customers are in their journey, businesses can provide targeted recommendations that increase the likelihood of converting potential... 

To maintain adaptability, businesses should anticipate changes in their environment. The commonly employed forecasting method within corporate circles is the bottom-up approach, which relies on historical data for projecting future trends. However, research suggests that this approach often falls short of accurately reflecting real-world events. This has led to the adoption of dynamic systems modeling, a technique grounded in the assumption of stable conditions. This method effectively replicates the system's current state, thereby assisting in predicting future behaviors over a longer timeframe. The dynamic systems modeling approach was employed in this study, underscoring the imperative... 

Providing quality hospital services depends on the efficient and correct implementation of processes. Therapeutic care processes are a set of activities that are carried out with the aim of diagnosing, treating and preventing any disease in order to improve and promote the patient's health. The purpose of this study is to use process mining techniques to discover and improve healthcare processes. The case study of this research is a psychiatric hospital in Shiraz. The approach implemented in this research consists of three main stages including data pre-processing, model discovery phase, and analysis phase. Three algorithms including Heuristic Miner, Inductive Miner, and ILP Miner were used... 

In the competitive world of software development, IT product development teams face numerous challenges and issues within business processes. In recent years, the combination of process mining and machine learning algorithms to optimize software development processes has gained attention. However, there is still a lack of a comprehensive model for identifying and conducting root-cause analysis of problems and providing solutions based on them in agile software development systems. The aim of this research is to find a solution for identifying and addressing issues in the software development process by modifying the change request workflow and examining various aspects of incidents. To... 

This thesis abstract addresses the use of process mining techniques when event data is stored at varying levels of granularity. While most techniques assume that events have the same level of granularity, real data is often stored differently. Pre-processing techniques allow for appropriate summarization of the data, which simplifies the output while retaining important process details. The goal is to ensure an interpretable output for stakeholders and different business teams without losing critical process points. However, adding new data as a feature to the dataset can be expensive, and at times, infeasible. Therefore, existing data is the only solution. To overcome this challenge, this... 

Dynamic portfolio optimization is one of the prominent problems in financial mathematics, for which numerous theories have been proposed to solve it. One of the solutions to this problem is the use of reinforcement learning. The main challenge with this method is that most reinforcement learning algorithms require a large amount of data, and therefore, the necessary data is often obtained not from the real world, but through simulations by estimating the parameters of a model. However, the approximation error of the parameters can propagate through the final solution, leading to inaccurate results. One approach to addressing this issue is the use of robust Markov processes and robust... 

This research focuses on examining and improving healthcare systems, particularly during crises and pandemics. Following disasters such as natural calamities and pandemics like COVID-19, healthcare systems face significant challenges due to the increased demand for medical services, creating a substantial threat to the population in the affected regions. This study emphasizes the importance of utilizing modern technologies such as the Internet of Things (IoT) and telemedicine systems in alleviating the pressure on healthcare systems. A combined approach of prediction and multi-objective optimization based on simulation is proposed in this study to improve resource allocation and demand... 

Deviations in business processes consistently impose significant financial and temporal costs on business owners and can lead to decreased customer satisfaction with organizations. Therefore, timely identification of deviations is a crucial and significant issue for business process managers. While extensive research has been conducted on the detection of antecedent deviations, predicting deviations before they occur—which could facilitate preemptive actions to prevent these deviations—has received less attention. In this context, the aim of this study is to predict two types of process deviations—temporal deviations and Rework deviations—using machine learning and deep learning algorithms,...