Short-term prediction of vehicle waiting queue at ferry terminal based on machine learning method

Ferry service plays an important role in several cities with waterfront areas. Transportation authorities often need to forecast volumes of vehicular traffic in queues waiting to board ships at ferry terminals to ensure sufficient capacity and establish schedules that meet demand. Several previous studies have developed models for long-term vehicle queue length prediction at ferry terminals using terminal operation data. Few studies, however, have been undertaken for short-term vehicular queue length prediction. In this study, machine learning methods including the artificial neural network (ANN) and support vector machine (SVM) are applied to predict vehicle waiting queue lengths at ferry terminals. Through time series analysis, the existence of a periodic queue-length pattern is established. Hence, methodologies used in this study take into account periodic features of vehicle queue data at terminals for prediction. To further consider the cyclical characteristics of vehicle queue data at ferry terminals, a prediction approach is proposed to decompose vehicle waiting queue length into two components: a periodic part and a dynamic part. A trigonometric regression function is introduced to capture the periodic component, and the dynamic part is modeled by SVM and ANN models. Moreover, an assembly technique for combining SVM and ANN models is proposed to aggregate multiple prediction models and in turn achieve better results than could be attained from a lone predictive method. The prediction results suggest that for multi-step ahead vehicle queue length prediction at ferry terminals, the ensemble model outperforms the separate prediction models and the hybrid models, especially as prediction step size increases. This research has important practical significance to both traffic service management interests and the travelers in cities along waterfront areas.     

A study on TLP hull sizing by utilizing optimization algorithm

This paper reports a study on hull sizing for tension leg platforms (TLPs) using an optimization model. At the conceptual or basic design stage of an offshore platform project, hull sizing is one of the most important tasks, having a great impact on later stages of the project. Hull sizing is usually done by an engineering team combining several disciplines and taking account of factors including hydrodynamics, global performance, stability, and structure. This makes it difficult and time-consuming to optimize the hull dimensions. In this study, hull sizing was modeled as an optimization problem in which a range of design criteria had to be satisfied and an objective function minimized. Sizing was then performed using optimization algorithms. For each module, the result calculated by our method was compared with that produced by commercial software (DNV Sesam). Our system was tested for conventional TLP design in three specific metocean environments. Finally, our optimized hull shape was compared with an existing TLP, and the difference in hull shape demanded by each environment was analyzed.     

Parametric study on the effects of pile inclination angle on the response of batter piles in offshore jacket platforms

Offshore jacket-type platforms are attached to the seabed by long batter piles. In this paper, results from a finite element analysis, verified against experimental data, are used to study the effect of the pile’s inclination angle, and its interaction with the geometrical properties of the pile and the geotechnical characteristics of the surrounding soil on the behavior of the inclined piles supporting the jacket platforms. Results show that the inclination angle is one of the main parameters affecting the behavior of an offshore pile. We investigated the effect of the inclination angle on the maximum von Mises stress, maximum von Mises elastic strain, maximum displacement vector sum, maximum displacement in the horizontal direction, and maximum displacement in the vertical direction. The pile seems to have an operationally optimal degree of inclination of approximately 5°. By exceeding this value, the instability in the surrounding soil under applied loads grows extensively in all the geotechnical properties considered. Cohesive soils tend to display poorer results compared to grained soils.     

Hydrodynamic performance of multiple-row slotted breakwaters

This study examines the hydrodynamic performance of multiple-row vertical slotted breakwaters. We developed a mathematical model based on an eigenfunction expansion method and a least squares technique for Stokes second-order waves. The numerical results obtained for limiting cases of double-row and triple-row walls are in good agreement with results of previous studies and experimental results. Comparisons with experimental measurements of the reflection, transmission, and dissipation coefficients (C _R , C _T , and C _E ) for double-row walls show that the proposed mathematical model adequately reproduces most of the important features. We found that for double-row walls, the C _R increases with increasing wave number, kd, and with a decreasing permeable wall part, dm. The C _T follows the opposite trend. The C _E slowly increases with an increasing kd for lower kd values, reaches a maximum, and then decreases again. In addition, an increasing porosity of dm would significantly decrease the C _R , while increasing the C _T . At lower values of kd, a decreasing porosity increases the C _E , but for high values of kd, a decreasing porosity reduces the C _E . The numerical results indicate that, for triple-row walls, the effect of the arrangement of the chamber widths on hydrodynamic characteristics is not significant, except when kd<0.5. Double-row slotted breakwaters may exhibit a good wave-absorbing performance at kd>0.5, where by the horizontal wave force may be smaller than that of a single wall. On the other hand, the difference between double-row and triple-row vertical slotted breakwaters is marginal.     

Panama Canal expansion: fuel economy and logistical risk

Since cargo capacity increases faster than fuel consumption, the significantly larger capacity fleets which will accompany expansion of the Panama Canal will introduce additional fuel economies and cost savings. Enabling larger, more fuel-efficient vessels to carry cargo the entire distance from Asia to US east-coast ports allows vessel operators to realize significant and meaningful savings compared with the alternatives of using smaller Panamax vessels for the whole distance, or sending the cargo over the US land bridge by train or truck. Fuel savings are quantified along with the monetary savings based on various assumptions for the price of fuel. These savings are dramatic and will increase directly with the price of crude petroleum. Finally, microeconomic theory is deployed to determine how cost savings will be distributed between shipping customers and vessel operators.     

Modeling of ship maneuvering motion using neural networks

In this paper, Neural Networks (NNs) are used in the modeling of ship maneuvering motion. A nonlinear response model and a linear hydrodynamic model of ship maneuvering motion are also investigated. The maneuverability indices and linear non-dimensional hydrodynamic derivatives in the models are identified by using two-layer feed forward NNs. The stability of parametric estimation is confirmed. Then, the ship maneuvering motion is predicted based on the obtained models. A comparison between the predicted results and the model test results demonstrates the validity of the proposed modeling method.     

An elastic-plastic iceberg material model considering temperature gradient effects and its application to numerical study

To simulate the FPSO-iceberg collision process more accurately, an elastic-plastic iceberg material model considering temperature gradient effects is proposed and applied. The model behaves linearly elastic until it reaches the ‘Tsai-Wu’ yield surfaces, which are a series of concentric elliptical curves of different sizes. Decreasing temperature results in a large yield surface. Failure criteria, based on the influence of accumulated plastic strain and hydrostatic pressure, are built into the model. Based on published experimental data on the relationship between depth and temperature in icebergs, three typical iceberg temperature profiles are proposed. According to these, ice elements located at different depths have different temperatures. The model is incorporated into LS-DYNA using a user-defined subroutine and applied to a simulation of FPSO collisions with different types of iceberg. Simulated area-pressure curves are compared with design codes to validate the iceberg model. The influence of iceberg shape and temperature on the collision process is analyzed. It is indicated that FPSO structural damage not only depends on the relative strength between the iceberg and the structure, but also depends on the local shape of the iceberg.     

Effect of pH on mechanical,physical and tribological properties of electroless Ni-P-Al<Subscript>2</Subscript>O<Subscript>3</Subscript> composite deposits for marine applications

Successful co-deposition of fine particulate matter within an Electroless Nickel-Phosphorous (ENi-P) matrix is dependent on various factors like bath composition, particle compatibility with metallic matrix, bath reactivity (pH), particle size and their distribution. ENi-P deposits incorporating Al₂O₃/Alumina in a disperse phase have varied effects on properties and attributes like surface roughness (Ra), microhardness, wear resistance, corrosion resistance and surface morphology of the deposits obtained. This paper experimentally investigates the effect of alumina (1.55 g/L) on Ra, microhardness, surface morphology, deposition rate, wettability, wear resistance and corrosion resistance of ENi-P-Al₂O₃ composite deposits on mild steel substrates at bath pH 5, 7 and 9. Study reveals that optimum deposit parameters and deposition rates are achieved with bath pH. However, not much study has been undertaken concerning composite deposits obtained from higher bath pH or basic bath. This is attributable to the fact that at higher bath pH or alkaline baths, the bath gets unstable and eventually degrades or decomposes, thereby resulting in sub optimal or poor deposition. Hence, experimental investigations carried out by preparing suitable baths, operating under optimum conditions, and enabling successful composite deposition in acidic and alkaline baths have revealed that there is a significant improvement in the above mentioned properties of the as-deposited composite deposits, as the pH is increased from pH 5 to pH 9. This aspect can therefore be advantageously utilized for preparing various marine components like fasteners, nuts, bolts, washers, pipes, cables, components having relative motion etc.     

Integrating land use and transport knowledge in strategy-making

There is broad and growing consensus regarding the central place of integrated Land Use and Transport (LUT) strategy development in establishing more efficient and sustainable urban environments. However, empirical evidence shows that such integration is hard to achieve in daily planning practice, due to many institutional barriers and substantive differences. More specifically, the tools developed to support LUT strategy development have very low implementation rates in daily planning practice. This paper introduces the concept of ‘knowledge generation’ as a potentially useful mechanism for closing the gap between support tools and planning practice. Through two specific Dutch planning cases, we analyze the applicability of this concept in supporting integrated LUT strategy development. The paper focuses on the developed strategies, how these differ from current practice, and how knowledge generation supported their development. We argue that socialization produces shared strategies and that effective socialization needs to be supported by efficient mutual exchange between tacit and explicit knowledge. We conclude by discussing the implications of this argument for the wider practice of LUT planning integration.     

Generation of synthetic datasets for discrete choice analysis

Despite the widespread use of synthetic data in discrete choice analysis, little is known about how the methodology used to generate synthetic datasets influences the properties of parameter estimates and the validity of results based on these estimates. That is, there are two potential sources of biases when using synthetic discrete choice data: (1) bias due to the method used to generate the dataset; and, (2) bias due to parameter estimation. The primary objective of this study is to examine bias due to the underlying data generation method. This study compares three methods for generating synthetic datasets and uses design of experiments and analysis of variance methods to investigate the ability to recover estimates for “true” logsum parameters for nested logit models. The method that uses nested logit probabilities to generate the chosen alternative results in unbiased parameter estimates. The method that is based on Gumbel error component approximations reveals that while the error components themselves are unbiased, subtle empirical identification problems can arise when these error components are combined with synthetically generated utility functions. The method that is based on normal error component approximations reveals that all logsum coefficients are biased upwards; the bias dramatically increases for those nests that have a low choice frequency and is most pronounced for those nests with high correlations among alternatives. Based on the results of the analysis, several recommendations for the generation of synthetic datasets for discrete choice analyses are provided.