Sharif Digital Repository

Utilizing the Axiomatic Design (AD) principles to develop a perfect product, design of a manufacturing system with minimal complexity is required. For the purpose of reducing the manufacturing system complexity, theoretically, it is preferred to integrate multiple Process Variables (PVs) of the product into a single process unit. However, due to significant presence of some active noise factors, this integration practice may result in failing to maintain the independence among some of Functional Requirements (FRs) of the product. This event is the result of statistical causal relationships unintentionally developed among a subset of the integrated PVs. In such a condition, the AD's... 

We consider the problem of learning causal models from observational data generated by linear non-Gaussian acyclic causal models with latent variables. Without considering the effect of latent variables, the inferred causal relationships among the observed variables are often wrong. Under faithfulness assumption, we propose a method to check whether there exists a causal path between any two observed variables. From this information, we can obtain the causal order among the observed variables. The next question is whether the causal effects can be uniquely identified as well. We show that causal effects among observed variables cannot be identified uniquely under mere assumptions of... 

The main goal in many fields in the empirical sciences is to discover causal relationships among a set of variables from observational data. PC algorithm is one of the promising solutions to learn underlying causal structure by performing a number of conditional independence tests. In this paper, we propose a novel GPU-based parallel algorithm, called cuPC, to execute an order-independent version of PC. The proposed solution has two variants, cuPC-E and cuPC-S, which parallelize PC in two different ways for multivariate normal distribution. Experimental results show the scalability of the proposed algorithms with respect to the number of variables, the number of samples, and different graph... 

Ways to reduce an airline's cost has been studied for years. In fact, in order to achieve this goal, ones need to know different types of airline's costs including; fuel and oil, maintenance, Ticketing, passenger services, etc. and their impact on the total cost of an airline. According to announcement of ICAO [1], "maintenance cost (about 11% of total cost) is the second major cost after fuel and oil". Therefore, this may attract airline owners' attention to find ways to control maintenance cost and eventually the total cost of an airline. Though, there are many systematic approaches to analyze cost reduction and making policies, in this article, SD modeling is applied to develop a... 

Brain connectivity estimation is a useful method to study brain functions and diagnose neuroscience disorders. Effective connectivity is a subdivision of brain connectivity which discusses the causal relationship between different parts of the brain. In this study, a dual Kalman-based method is used for effective connectivity estimation. Because of connectivity changes in autism, the method is applied to autistic signals for effective connectivity estimation. For method validation, the dual Kalman based method is compared with other connectivity estimation methods by estimation error and the dual Kalman-based method gives acceptable results with less estimation errors. Then, connectivities... 

We study the problem of experiment design to learn causal structures from interventional data. We consider an active learning setting in which the experimenter decides to intervene on one of the variables in the system in each step and uses the results of the intervention to recover further causal relationships among the variables. The goal is to fully identify the causal structures with minimum number of interventions. We present the first deep reinforcement learning based solution for the problem of experiment design. In the proposed method, we embed input graphs to vectors using a graph neural network and feed them to another neural network which outputs a variable for performing...