Abstract
More and more frequently, the need of vibration mitigation induced by railways, tramways and metros demands for the use of complex trackforms, characterised by innovative materials, components and laying technologies. Whilst on the one hand, we have recently gained useful standardised tools for the test of track components (see the EN series 1323 0 and 134 81), on the other hand, we still suffer from a shortage of standards to test and design the whole trackform. In UNI (Italian Standardisation Body) we have tried to fill this shortage, developing a fully-comprehensive standard for the assessment of the static and dynamic behaviour of railway/metro/tramway track systems. The standard covers the general criteria for the design, qualification and verification on service of a trackform. The standard is composed by 4 parts. Part 1 contains a description of the constitutional components of a trackform and defines the parameters to be considered. Part 2 provides the criteria for a mathematical modelling. Part 3 and Part 4 provide respectively the procedures for laboratory and in-situ testing. This paper summarises the main characteristics of this new standard, with part of the choices that have brought to the final result.

Abstract
In this paper, 4 models of average stress states have been calculated on the base of focal mechanism data and 25 focal mechanisms of the largest earthquakes belong to different segments of Manila subduction zone and adjacent faults are chosen. In order to identify the recent movement pattern in the studied fault systems, some consent criteria of classifying focal mechanisms and average stress states have been drawn out. Based on comparative analysis of the special correlation between the stress distribution patterns with kinematic-geometric parameters of faults, characteristics of average tectonic stress field and recent tectonic movements have been defined for the systems/or segments of the active faults in the studied region.

Abstract
In the UK the accepted method of preventing the migration of fines from the subgrade soil into the ballast is to place a 100 mm thick layer of sand overlain by a geotextile at the interface of the ballast and subgrade. However, the placement of sand in significant thickness is resource hungry. Thus a number of geotextile manufacturers have proposed products that will replace the sand layer with thinner and easy to place composites. In order to prove that these composites are likely to work in situ a test has been developed in conjunction with Network Rail, the owner and operator of Britains railway network. The test is reported here, together with results of some of the investigations carried out to develop a thinner composite that is easier to place and is as at least as effective as the thicker sand/textile layer that is used currently used.

Abstract
The critical first step in the seismic event processing is to identify the class of an unknown detected event. An essential discrimination task is to allocate the incoming event to a group associated with the kind of physical phenomena producing it. This paper proposes a neural network architecture for automatic seismic discrimination between earthquakes and quarry blast. Such architecture, called Fuzzy ARTMAP, combines the salient properties of fuzzy set with adaptive resonance theory to make a powerful network. A map field is added to perform supervised learning. Fuzzy ARTMAP has certain advantages over many other network models and is more suited for classification problem. These characteristics make it an attractive neural network model for investigation into the problem of seismic event discrimination. To reduce the complexity of the neural network model and time computation, important features are derived from seismic signals using signal processing methods. Discrimination results show that Fuzzy ARTMAP deal to favorable levels of learned accuracy, speed and generation even when the amount of training data is limited. Furthermore, Fuzzy ARTMAP provides incremental, stable and on-line learning.