Search

Falls are the leading cause of fatal injuries in the elderly such as fractures, and secondary damage from falls can lead to death. As such, fall detection is a crucial topic. However, due to the trade-off relationship between privacy preservation, user convenience, and fall detection performance, it is generally difficult to develop a fall detection system that simultaneously satisfies all conditions. The main goal of this study is to build a practical fall detection framework that can effectively classify the various behavior types into &ldquo, Fall&rdquo, and &ldquo, Activities of daily living (ADL)&rdquo, while securing privacy preservation and user convenience. For this purpose, signal data containing the motion information of objects was collected using a non-contact, unobtrusive, and non-restraint impulse-radio ultra wideband (IR-UWB) radar. These data were then applied to a convolutional neural network (CNN) algorithm to create an object behavior type classifier that can classify the behavior types of objects into &ldquo, ADL.&rdquo, The data were collected by actually performing various activities of daily living, including falling. The performance of the classifier yielded satisfactory results. By combining an IR-UWB and CNN algorithm, this study demonstrates the feasibility of building a practical fall detection system that exceeds a certain level of detection accuracy while also ensuring privacy preservation and user convenience.

The integrated circuits (ICs) on wafers are highly vulnerable to defects generated during the semiconductor manufacturing process. The spatial patterns of locally clustered defects are likely to contain information related to the defect generating mechanism. For the purpose of yield management, we propose a multi-step adaptive resonance theory (ART1) algorithm in order to accurately recognise the defect patterns scattered over a wafer. The proposed algorithm consists of a new similarity measure, based on the p-norm ratio and run-length encoding technique and pre-processing procedure: the variable resolution array and zooming strategy. The performance of the algorithm is evaluated based on the statistical models for four types of simulated defect patterns, each of which typically occurs during fabrication of ICs: random patterns by a spatial homogeneous Poisson process, ellipsoid patterns by a multivariate normal, curvilinear patterns by a principal curve, and ring patterns by a spherical shell. Computatio...

In this research report, we propose a heuristic algorithm with some primal recovery strategies for capacitated facility location problems (CFLP), which is a well-known combinatorial optimization problem with applications in distribution, transportation and production planning. Many algorithms employ the branch-and-bound technique in order to solve the CFLP. There are also some different approaches which can recover primal solutions while exploiting the primal and dual structure simultaneously. One of them is a MVCD (Mean Value Cross Decomposition) ensuring convergence without solving a master problem. The MVCD was designed to handle LP-problems, but it was applied in mixed integer problems. However the MVCD has been applied to only uncapacitated facility location problems (UFLP), because it was very difficult to obtain "Integrality" property of Lagrangian dual subproblems sustaining the feasibility to primal problems. We present some heuristic strategies to recover primal feasible integer solutions, handling the accumulated primal solutions of the dual subproblem, which are used as input to the primal subproblem in the mean value cross decomposition technique, without requiring solutions to a master problem. Computational results for a set of various problem instances are reported.

The traveling salesman problem with precedence constraints (TSPPC) is one of the most difficult combinatorial optimization problems. In this paper, an efficient genetic algorithm (GA) to solve the TSPPC is presented. The key concept of the proposed GA is a topological sort (TS), which is defined as an ordering of vertices in a directed graph. Also, a new crossover operation is developed for the proposed GA. The results of numerical experiments show that the proposed GA produces an optimal solution and shows superior performance compared to the traditional algorithms.

In this paper, we present variants of Shor and Zhurbenko's r-algorithm, motivated by the memoryless and limited memory updates for differentiable quasi-Newton methods. This well known r-algorithm, which employs a space dilation strategy in the direction of the difference between two successive subgradients, is recognized as being one of the most effective procedures for solving nondifferentiable optimization problems. However, the method needs to store the space dilation matrix and update it at every iteration, resulting in a substantial computational burden for large-sized problems. To circumvent this difficulty, we first propose a memoryless update scheme, which under a suitable choice of parameters, yields a direction of motion that turns out to be a convex combination of two successive anti-subgradients. Moreover, in the space transformation sense, the new update scheme can be viewed as a combination of space dilation and reduction operations. We prove convergence of this new method, and demonstrate how it can be used in conjunction with a variable target value method that allows a practical, convergent implementation of the method. We also examine a memoryless variant that uses a fixed dilation parameter instead of varying degrees of dilation and/or reduction as in the former algorithm, as well as another variant that examines a two-step limited memory update. These variants are tested along with Shor's r-algorithm and also a modified version of a related algorithm due to Polyak that employs a projection onto a pair of Kelley's cutting planes. We use a set of standard test problems from the literature as well as randomly generated dual transportation and assignment problems in our computational experiments. The results exhibit that the proposed space dilation and reduction method and the modification of Polyak's method are competitive, and offer a substantial advantage over the r-algorithm and over the other tested limited memory variants with respect to accuracy as well as effort.

Subgradient optimization methods provide a valuable tool for obtaining a lower bound of specially structured linear programming or linear programming relaxation of discrete optimization problems. However, there is no practical rule for obtaining primal optimal solutions from subgradient-based approach other than the lower bounds. This paper presents a class of procedures to recover primal solutions directly from the information generated in the process of using subgradient optimization methods to solve such Lagrangian dual formulations. We also present a hybrid primal dual algorithm based on these methods and some computational results.

In this article, we propose a practical algorithm for capacitated facility location problems (CFLP). There are some approaches which can obtain primal solutions while simultaneously exploiting the primal structure and the dual structure. One of these approaches is the mean value cross decomposition (MVCD) method that ensures convergence without solving master problems. However, MVCD has been previously applied only to uncapacitated facility location problems (UFLP), due to the fact that the performance is highly dependent on the structure of the problem. The proposed algorithm, named the dynamic mean value cross decomposition algorithm (DMVCD), is effectively integrated with MVCD and cutting plane methods in order to tighten the bounds by reducing the duality gap. Computational results of various instances are also reported to verify the effectiveness and efficiency of DMVCD.