In this paper, a genetic algorithm is presented for the simultaneous layout and pipe size optimization of water distribution networks and its eciency is evaluated and compared using dierent selection mechanisms. An engineering concept of reliability is used, in which the number of independent paths from source nodes to each of the demand nodes is considered as a measure of reliability. The method starts with a predened layout, which includes all possible links. The method is capable of designing a layout of predened reliability, including tree-like and looped networks. It is shown that a layout optimization of a network, followed by size optimization, does not lead to an optimal or a near-optimal solution. This emphasizes the need for simultaneous layout and size optimization of networks, if an optimal solution is desired. The performance of the method for layout optimization of pipe networks is tested against two benchmark examples in the literature and the results are presented.

In this paper, the Markov models, eigenvectors and eigenvalue concepts are used to propose a methodology for analyzing the transient reliability of a system with identical components and identical repairmen. The components of the systems under consideration can have two distinct congurations, namely

they can be arranged in series or in parallel. A third case is also considered, in which the system is up (good) if k-out-of-n components are good. For all three cases, a procedure is proposed for calculating the transient probability of the system availability and the duration of the system to reach the steady state.

In this paper, an analytical method is developed to compute the project completion time distribution in a dynamic PERT network, where the activity durations are exponentially distributed random variables. The projects are generated according to a renewal process and share the same facilities. Thus, these projects cannot be analyzed independently. The authors' approach is to transform this dynamic PERT network into a stochastic network and, then, to obtain the project completion time distribution by constructing a proper continuous-time Markov chain. This dynamic PERT network is represented as a network of queues, where the service times represent the durations of the corresponding activities and the arrival stream to each node follows a renewal process. Finally, the proposed methodology is extended to the generalized Erlang activity durations.

In this paper, rst, the equations of motion for a rectangular isotropic plate have been derived. This derivation is based on the Von Karmann theory and the eects of shear deformation have been considered. Introducing an Airy stress function, the equations of motion have been transformed to a nonlinear coupled equation. Using the Galerkin method, this equation has been separated into position and time functions. By means of the dimensional analysis, it is shown that the orders of magnitude for nonlinear terms are very small, with respect to linear terms. Then, for the rst time, the invariant manifold theory has been applied to the plate problem and it is proved that the nonlinearities are stiness and inertia types. Finally, the multiple scale method is applied to the equations of motion and closed-form relations for the nonlinear natural frequencies and mode shapes of the plate are derived. The obtained results are in good agreement in comparison with numerical methods. Using the obtained relation, the eects of initial displacement, thickness and dimensions of the plate on nonlinear natural frequencies and displacements have been investigated. These results are valid for a special range of the ratio of thickness to dimensions of the plate, which is a characteristic of the multiple scale method.

In this investigation, a computational model has been developed, including heat transfer and the eects of the resistance of a metal-mold interface and pressure, for simulation of the solidication process. Simulation of the interface resistance is based on the Zero Thickness Element (ZTE), utilizing the Finite Element Method (FEM). Solid boundary conditions, including contact resistances, have been modied by a pressure gradient in each of the ZTE. The pressure gradient has been modeled, based on experimental data. In order to verify the computational results, an Al-11.5 wt% Si alloy was poured into a permanent mold and the temperature of the interface was measured by a data acquisition system. Then, the eect of metalo-static pressure on overall heat transfer in the interface resistance was modeled. Comparison between the experimental and simulation results during the solidication process shows a good consistency, which conrms the accuracy of the model for the eects of interface resistance on solidication time.

The objective of this paper is to introduce a new direct method for neutronic calculations. This method, called Direct Discrete Method (DDM), is simpler than the Neutron Transport Equation and more compatible with the physical meanings of the problem. The method, based on the physics of the problem, initially runs through meshing of the desired geometry. Next, the balance equation for each mesh interval is written. Considering the connection between the mesh intervals, the nal discrete equation series are directly obtained without the need to rst pass through the set-up of the neutron transport dierential equation. In this paper, a single and multigroup neutron transport discrete equation has been produced for a cylindrical shape fuel element with and without the associated clad and the coolant regions, each with two dierent external boundary conditions. The validity of the results from this new method are tested against the results obtained by the MCNP-4B and the ANISN codes.

Application of the base-isolation systems, as a means to limit the seismic-induced response of structures, has attracted the attention of many engineers and researchers. Due to their importance, the Uniform Building Code (UBC) has incorporated a special section for the seismic analysis and design of base-isolated structures since its 1991 edition. The present work investigates the vertical distribution of the lateral seismic force for base-isolated structures provided by the 1997 edition of UBC (UBC97). Dierent 6 and 8-story, 3-D base-isolated structural models with LRB isolators are considered, having a variety of eective periods and eective damping ratios. The UBC97 analysis procedure for the base-isolated structures is used to determine the minimum lateral seismic force and its vertical distribution for dierent oors. Since the number of stories above the isolation interface is more than four for the considered isolated structural models, the response spectrum analysis is used, considering the equivalent linear properties for isolation systems. Also, the UBC97 recommended that the 5%-damped design spectra be properly modied to account for the actual modal damping ratios of an isolated structure. Extensive nonlinear dynamic analyses were performed for 8 types of LRB isolators, using appropriately normalized earthquake accelerograms recorded on SA and SB soil proles. Both the superstructure and the isolators are allowed to behave nonlinearly, in order to evaluate the seismic induced demand shear force on dierent oors. The peak base center and corner displacements, maximum base and story shear forces and the maximum inter-story drifts are determined for dierent base-isolation systems and earthquake records. The results, together with their mean and mean plus one standard deviation values, are used for the evaluation of UBC97 response spectrum analysis procedures for these buildings. The results indicate that the UBC97 suitably predicts the seismic lateral forces for base-isolated buildings. However, it does not provide a good estimate of the shear force distribution over the height, especially for the highly damped base-isolation systems. Furthermore, the number of columns per story that behaved nonlinearly during the time history analyses is included for comparison.

This paper describes an attempt to develop congestion indices for a developing country with a limited sampling. Speed and ow rate information was collected from urban highway trac ow videotaping in Iran. Video display provided information for trac modeling when, similar to most of the developing nations, Iran does not have locally developed highway capacity manuals. Processing of the extracted data and trac modeling provided free and capacity ow information for the selected freeway and principal arterial sections for the capital city of Tehran. The congestion indices were associated, in simple and logical models, with ve trac descriptive variables of travel speed, travel rate, delay rate, travel rate ratio and delay ratio, respectively. To calibrate the congestion index models, specic congestion levels were assessed by the index values under free and capacity ow conditions. The functional forms used in congestion index modeling were derived from the congestion rate of change, simple, reasonable and relevant assumptions. For comparison, index models, based on the US Highway Capacity Manual, HCM, speed and ow rate information, were also developed. The developed indices were suggested as feasible and included quick response measures for congestion monitoring and control when trac management resources are often limited and scarce. Based on the preferences of a group of transportation professionals and university students, the relative importance of each of the individual indices, as average weighting value, was identied. Using the average weighting values, an overall congestion index was also developed. The overall congestion index was suggested as a complimentary or alternative measure for congestion intensity assessment and evaluation. The study ndings were based on a rather limited database and were location specic

nevertheless, for quick response congestion index development, the applied methodology could be used by any developing country. The suggested indices have potential in trac management and congestion mitigation.

In this paper, the modeling and impedance-control of a one link exible robot is presented. The concept of impedance control of exible link robots is rather new and is being addressed for the rst time. The control algorithm is valid for both constrained and unconstrained maneuvers. First, equations of motion and the associated boundary conditions are derived using Hamilton's principle. A linear nite dimensional model is, then, generated in the Cartesian coordinates, using the assumed mode method and by introduction of a proper coordinate transformation. The target impedance is, then, introduced in the Cartesian coordinate system and a control law is designed to realize the proposed target impedance for a given frequency range, using the Sliding Mode Control Theory. A set of computer simulations are carried out to demonstrate the eectiveness of the proposed control law. Simulations are carried out with various contact stiness. As the results show, when the environmental surface stiness is smaller than, or comparable to, that of the link, the control system is able to achieve stable behavior and the link vibration diminishes rather rapidly. However, when the environmental stiness is much greater than that of the stiness of the link, although the robot achieves stable behavior during contact, the vibrations tend to increase.