Horizontal Compaction Research Articles

An improved small-scale test setup for assessing backward erosion piping

Backward erosion piping is one of the erosion mechanisms that lead to embankment dam incidents, whereby foundation granular materials are transported to the downstream toe leaving a shallow pipe in place. Most previous studies have focused on secondary erosion as the responsible process for the pipe progression. On the other hand, a few experimental investigations have considered primary erosion by restricting the pipe pathway within a confined channel for easier monitoring of local gradients. This paper presents a modified small-scale setup with the hole-type exit configuration to capture local hydraulic conditions associated with the progression of the pipe in a realistic and unconstrained pathway. Horizontal compaction and a more accurate measurement and acquisition system are other improvements added to this setup. The test outcomes of the two fine sands have revealed that the scale factor of Sellmeijer’s model can be used to predict the critical global hydraulic gradient for the specific geometry of the small-scale setup with the hole-type exit. Moreover, the progression of the pipe remains unaffected by loading history. Results also confirm the previous findings regarding the mutual influence of local gradient and pipe tip progression.

Evaluating the efficiency of stone columns in mitigating liquefaction risks and enhancing soil bearing capacity: A case study

Several processes allow the improvement of soils such as vibrocompaction, stony columns, static horizontal compaction etc. These processes reduce the risk of liquefaction potential and making it possible to build on this type of soil when the space restriction require it. Stone columns are a recognized method of soil improvement, which consists of creating large diameter columns using special vibrators with granular filling materials introduced into the ground. The objective of this work is to evaluate the effectiveness of stone columns, made in a seismic zone containing liquefiable materials, with regards to the reduction of the risk of liquefaction and the improvement of the bearing capacity of the soil. The approach followed is the exploitation of geotechnical investigation tests (CPT Cone Penetration Test), (SPT: Cone Penetration Test, (pressuremeter tests), carried out before and after soil treatment. This study showed that the network of gravelled columns produces an enhanced soil improvement effect by improving the modulus of elasticity of the soil and reducing settlement and the risk of liquefaction after treatment. The process allows for the improvement of the geotechnical characteristics of its soils, making it possible to build infrastructure and development projects in the area.

Comparison of the mass tissue strength of strawberry fruit between vertical and horizontal compaction

Strawberries have very thin skin that is prone to damage during post-harvest handling. The strength of this fruit mass tissue has a difference between vertical and horizontal directions. Therefore, this study aimed to compare the strength of the strawberry mass tissue between vertical and horizontal directions using a compressive test at different speeds and compressibility levels. The research was conducted by placing a load on the sample from the vertical and horizontal directions. The variations in compression speed were 2, 4, and 6 mm/s, and the compressibility levels were 6, 12, 18, 24, and 30%, respectively. The results showed that the compacting from the vertical direction obtained a combined mechanical response between the fruit structure and mass tissue cells, while from the horizontal direction the mechanical response was only obtained from the fruit structure. Along with the duration of the compacting, mass tissue damage has started to occur. In this phase, the fruit cells begin to break down in the skin causing the mass tissue damage to decrease. The results of this study can be used to develop a more efficient and effective packaging model for strawberries that has never been disclosed before.

Impact of horizontal soil strain on flexible manhole riser deflection based on laboratory test results

Underground mining extraction causes deformation of the rock mass and the surface, which may cause damage to the construction of manholes, thus leading to the failure of the sewage system. To ensure the proper operation of sewer systems in mining areas, it is important to consider the interaction between the manhole and the horizontally strained soil. The impact of continuous mining deformations of the ground around the manhole manifests itself through the impact of horizontal strains, which mainly affect the manhole riser. These strains cause the horizontal cross-sectional deflection of the flexible manhole riser, which is made of thermoplastics.This article presents the results of laboratory (model) tests conducted in conditions similar to in-situ conditions. The aim of these studies was to determine the impact of horizontal strains of the non-cohesive soil on the relative deflection of the horizontal cross-section of the manhole riser, which was made of thermoplastic. The laboratory tests have been performed for manhole riser models with various ring stiffnesses, which are used in practice, and at various manhole foundation depths. The values of the relative cross-sectional deflection of the manhole riser models have been determined for both the horizontal loosening and horizontal compaction of the soil. The research methodology has been presented, and the results of laboratory tests have also been analysed. The horizontal cross-sectional deflection of the flexible manhole riser depends not only on the values of the horizontal soil strain but also on the ring stiffness and the foundation depth of the discussed cross-section of the riser.

Structure and Deformation History of the Rapidly Growing Tainan Anticline at the Deformation Front of the Taiwan Mountain Belt

AbstractThis study aims at further documenting the mechanisms of shortening at the front of fold‐and‐thrust belts. We focus on an actively growing anticline located at the deformation front of the Taiwan fold‐and‐thrust belt. Based on a multidisciplinary approach combining mainly subsurface data and geodetic techniques, we show that the Tainan anticline is a pure‐shear fault‐bend fold growing above a 38–45° west dipping back thrust, the Houchiali fault, rooted on a 3.8‐km‐deep detachment. The cumulative shortening is estimated at 2–3 km since 310 ± 50 ka, including ~30–50% of horizontal compaction shortening. The significance of the fold is little in terms of total shortening at the scale of the mountain piedmont, yet the Holocene shortening rate of 10.3 ± 1.0 mm/a accounts for 25% of the present‐day shortening rate across the piedmont. Earthquake scaling relationships applied to the Houchiali fault predict Mw~6 earthquakes that would occur a lot more frequently than indicated from historical earthquake catalogs. Hence, the aseismic slip behavior observed from geodetic measurements since two decades is a representative behavior of the fault at least at the scale of a few centuries. Our results bear out the dominance of pure‐shear folding at the front of fold‐and‐thrust belts and support horizontal compaction as a significant shortening mechanism. In contrast, the back thrust wedge structure and the aseismic slip are peculiar characteristics that likely arise from the combination of low friction and high‐pore pressure related to the thick mudstone formation hosting the wedge and of high syntectonic sedimentation rates.

Crystal Mush Flow in Small Concentrically Expanded Pluton (Castle Crags Pluton; Klamath Mountains, CA, USA)

AbstractUnderstanding of the processes of magmatic fabric formation in crystal‐rich magmas and their reflection in rock magnetic properties are important for understanding pluton formation and intrusion mechanisms. On the example of small concentrically zoned Castle Crags pluton in the Klamath Mountains (CA, USA) we provide reconstruction of the flow/deformation mechanisms of the crystal‐rich magma and pluton growth based on detailed structural mapping and microstructural analysis employing the anisotropy of magnetic susceptibility, microstructural analysis, and crystallographic preferred orientation. Our study reveal microstructural evidence for progressive development of magmatic textures in the pluton core transitioning to submagmatic and eventually subsolidus fabric at the pluton periphery, that is interpreted in terms of the flow/deformation of the crystal mush. The documented magmatic textures are linked to anisotropy of magnetic susceptibility parameters and orientation. The recorded anomalous degree of anisotropy of magnetic susceptibility found in the pluton is attributed to tiling and plastic deformation of magnetite grains by the surrounding phenocrysts. The concentric structure of the pluton resulted from horizontal compaction and margin parallel stretching of the dense crystal mush around the vertically intruding trondhjemite magma in the pluton core. The Castle Crags pluton is interpreted as a concentrically expanded pluton, which grew at least by two increments of granodiorite and trondhjemite magma emplacement.

Area-constant strain and dilation in sandbox models: Insights from whole-model area balance

Line-length restoration of sandbox models of contractional structures typically shows that bed lengths are too short, implying that the models are not area balanced. Two alternative hypotheses have been proposed to explain this discrepancy: horizontal (or tectonic) compaction, a porosity loss caused by horizontal shortening, and area-constant shortening with bed thickening. Whole-model area-balance shows that two very different sandbox models, for which line-length restoration implies compaction, are actually area balanced with small dilations. The line-length imbalance is due to area-constant thickening that accompanies the shortening. This thickening may not be obvious to the eye and may be impossible to recognize in the deformed state unless the initial thickness is known independently because when the entire model is shortened and thickened, no undeformed thickness may remain.

The importance of missing strain in Deep Water Fold and Thrust Belts

Deep water fold and thrust belts (DWFTBs) are sedimentary wedges that accommodate plate-scale deformation on both active and passive continental margins. Internally, these wedges consist of individual structures that strongly influence sediment dispersal, bathymetry and fluid migration. Most DWFTB studies investigate basin- and intra-wedge- scale processes using seismic reflection profiles, yet are inherently limited by seismic resolution. Of critical importance is strain distribution and its accommodation on discrete faults compared to distributed deformation. Recent studies have considered strain distribution by investigating regional reflection DWFTBs profiles within coupled systems, which contain down-dip compression and up-dip extension. There is broad agreement of a mis-balance in compression versus extension, with ∼5% excess in the latter associated with horizontal compaction, yet this remains unproven.Using two exceptionally well exposed outcrops in the Spanish Pyrenees we consider deformation of DWFTB at a scale comparable to, and beyond, seismic resolution for the first time. By coupling outcrop observations (decametre to hectometre scale) with a re-evaluation of seismic profiles from the Orange Basin, South Africa, which contains one of the best imaged DWFTBs globally, we provide a unique insight into the deformation from metre to margin scale. Our observations reveal hitherto unrecognised second order structures that account for the majority of the previously recognised missing strain. This re-evaluation implies that ∼5% missing strain should be accounted for in all DWFTBs, therefore existing studies using restorations of the sediment wedge will have underestimated crustal shortening in active margins, or sedimentary shortening in gravity driven systems by this amount. In contrast to previous studies, our observations imply that the majority of this strain is accommodated on discrete fault surfaces and this can explain the occurrence and location of a range of intra-wedge processes that are intimately linked to structures including sediment dispersal, fluid migration pathways and reservoir compartmentalisation.

A comparative study of conventional and controlled traffic in irrigated cotton: I. Heavy machinery impact on the soil resource

To increase in-field efficiency of mechanical operations, machines have increased in size and capacity, which has resulted in cotton pickers approaching 40Mg in mass. Such mass presents concern for soil compaction and subsequent crop performance. Controlled traffic farming (CTF) is a useful means to reduce machine impact, but is not highly adopted globally. This investigation was designed to compare a standard cotton picker side-by-side to a CTF converted cotton picker in terms of soil resource impact. A replicated, side-by-side, commercial scale experiment was instigated with a 1.5m row-spacing CTF system compared against a 1.0m row-spacing standard cotton system. Soil moisture, bulk density, and strength were measured immediately prior and post harvest, along with crop yield, for two cotton seasons and one wheat season. In depth cotton quality, water use efficiency and economic analysis are undertaken in a companion paper. Both machines were shown to induce comparable soil compaction, effects detectable to 0.8m soil depth, with the main difference being a 17% reduction in furrow traffic for the CTF system. The inner and outer dual wheel of the standard picking system had similar impact, effectively creating a horizontal compaction pan at ≈0.35–0.40m depth, perpendicular to the direction of machine travel. Bio-ripping of this compaction pan by wheat was demonstrated as ineffectual. The CTF system grew less cotton plants per hectare, but outperformed the standard system. Bio-ripping more effectively accessed water in the CTF system and out yielded the conventional system by ≈60%.

Fracture Alignments in Marine Sediments Off Vancouver Island fromPsSplitting Analysis

Alignments of fractures and cracks in marine sediments may be controlled by various mechanisms such as horizontal compaction and extension and basement faulting. The orientation of these alignments can be estimated through analyses of S‐wave splitting. If sensors in ocean‐bottom observations are deployed through free fall, sensor orientation needs to be determined in order for the recorded data to be used for such analyses. Here, we estimate the sensor orientation from the linear particle motions of P‐to‐s (Ps) phases converted at the sediment–basement interface and also from T waves that are excited by earthquakes and propagate in the seawater. We examine waveforms of local earthquakes recorded by 32 ocean‐bottom seismometers (OBSs) that were deployed through free fall for three months in 2010 off Vancouver Island where the strike‐slip Nootka fault zone (NFZ) intersects the deformation front of the Cascadia subduction zone. Because the particle motion of the Ps wave was corrected by estimating splitting parameters, the fast polarization direction, which reflects S‐wave anisotropic structure within the sediment, can also be evaluated. Consequently, we could estimate the fast polarization direction at OBSs deployed near the NFZ and west of the deformation front. The obtained fast directions appeared to correspond to alignments of shear fractures in the marine sediments associated with the left‐lateral motion of the fault in the basement along the NFZ, margin‐normal cracks due to horizontal compression west of and slightly away from the deformation front, and frontal thrust faults within the accretionary prism near the deformation front.

Compact visualization of Java program execution

The context of this work is a practical, open-source visualization system, called JIVE, that supports two forms of runtime visualizations of Java programs - object diagrams and sequence diagrams. They capture, respectively, the current execution state and execution history of a Java program. These diagrams are similar to those found in the UML for specifying design-time decisions. In our work, we construct these diagrams at execution time, thereby ensuring continuity of notation from design to execution. In so doing, a few extensions to the UML notation are proposed in order to better represent runtime behavior. As sequence diagrams can become long and unwieldy, we present techniques for their compact representation. A key result in this paper is a novel labeling scheme based upon regular expressions to compactly represent long sequences and an Or2 algorithm for computing these labels, where r is the length of the input sequence, based upon the concept of 'tandem repeats' in a sequence. Horizontal compaction greatly helps minimize the extent of white space in sequence diagrams by the elimination of object lifelines and also by grouping lifelines together. We propose a novel extension to the sequence diagram to deal with out-of-model calls when the lifelines of certain classes of objects are filtered out of the visualization, but method calls may occur between in-model and out-of-model calls. The paper also presents compaction techniques for multi-threaded Java execution with different forms of synchronization. Finally, we present experimental results from compacting the runtime visualizations of a variety of Java programs and execution trace sizes in order to demonstrate the practicality and efficacy of our techniques. Copyright © 2016 John Wiley & Sons, Ltd.

The interplay between deformation, erosion and sedimentation in the deep‐water Mexican Ridges foldbelt, western Gulf of Mexico basin

AbstractWe analyse a regional 2D seismic section of the Mexican Ridges foldbelt (MRFB), western Gulf of Mexico, and construct excess‐area diagrams for each of the structures comprising the foldbelt to estimate shortening, the onset of folding and the degradation of the folded seafloor. From the chronostratigraphy, we derive rates of tectonic and superficial mass transport and illustrate how they change across the MRFB. The resulting tectonic transport in the MRFB is 11.8 km forming a train of twelve buckle folds above a detachment at a depth of ca. 6 s of two‐way travel time, with an average strain of ca. 10%. The fold train grew at a mean uplift rate of ca. 0.21 mm year−1. Cross‐sectional balancing demonstrates that shortening balances the down‐slip motion of the Quetzalcoatl extensional system (QES), suggesting that horizontal compaction, volume loss and other penetrative deformation mechanisms are negligible. By assuming steady‐state denudation, we are able to distinguish sediments derived locally from sediments transported from distant sources. The constant of mass diffusivity, a parameter controlling the degradation rate, is ca. 0.42 m2 year −1, which is characteristic of rapid, episodic, superficial mass movements. The combined sedimentation rate from both, local and distal sources is ca. 0.23 mm year −1. Those values are not constant; structures proximal to the continental shelf are rising rapidly and are being degraded more intensely than those in the distal part of the MRFB, where sedimentation outweighs tectonic uplift. Our results indicate deformation initiated up to 3 Myr earlier than estimated from stacking patterns. Moreover, we find deformation started synchronously during the Late Miocene throughout the MRFB and not in two episodes as the stacking relations suggest. The discrepancy can be explained by a delay in the sedimentary response to folding. During early fold growth, nearly constant thickness strata are deposited before a progressive unconformity and other converging geometries develop. The development of growth strata is fast in the folds near the QES, which are being uplifted rapidly and degraded vigorously. Under these conditions, the stratigraphic relations give only a broad estimate of the pretectonic/syntectonic limit when compared to the excess‐area method. On the other hand, the development of growth strata took twice as much time for folds near the abyssal plain, which are being uplifted at a slower rate and where degradation is less intense. Consequently, the delay takes more time, and the use of stratigraphic relations introduces an even more pronounced bias towards younger ages in the identification of the onset of folding.

PROCESS MANAGEMENT IN RUSSIAN BUSINESS EDUCATION

The current management system in Russian business education is predominantly based on the principle of functional management, which in turn does not comprehensively address the emerging objectives and terms of modern-day education providers in Russia. The countries’ business schools are of the particular concern of the issue, considering the factors of their recent establishment and revenue-based financing. Presented in this paper is the outcome of implementation of the process-orientated approach to business education based on the experience of one of the leading Russian business schools. The implementation of the process-orientated approach into the management system of the business school has resulted in an array of advantages: a significant improvement in the speed of the management process; a horizontal and vertical compaction of processes due to employees making independent decisions, reduction in the number of errors, delays and alterations, an increase in the quality of tasks executed by the personnel, a decrease in the need of employee monitoring; reduction of costs associated with employee wages and equipping workplaces; development of the basis for automation of educational and supporting business processes and delegation of particular elements of business processes to external.

Coupled fluid flow and sediment deformation in margin‐scale salt‐tectonic systems: 2. Layered sediment models and application to the northwestern Gulf of Mexico

In paper 1 we described a methodology to model coupled fluid flow and deformation in composite salt and siliciclastic tectonic systems and investigated their compaction and overpressuring behavior prior to and during continental margin‐scale gravitational spreading. Compaction‐driven Darcy fluid flow in clastic sediments is coupled through the effective pressure to their frictional‐plastic yield and mechanical deformation. Viscous flow of the underlying salt is independent of fluid pressure. Paper 1 presented prototype models that are limited to single uniform sediment lithologies, either sandstone‐type or shale‐type, that undergo mechanical and volumetric viscous compaction. In this paper we present models with layered sandstone‐type and shale‐type lithologies designed to better approximate the more complex stratigraphy of the Gulf of Mexico, our natural example. A first set of models demonstrates that layered lithologies can produce fluid pressure regimes similar to those observed in sedimentary basins. We then introduce an improved formulation of viscous compaction that includes a stronger dependence on porosity and depth (used as proxy for temperature), thereby more effectively self‐limiting viscous compaction. A second set of models with the improved viscous compaction formulation demonstrates that the onset of gravity spreading is mainly controlled by overpressuring in the landward end of the salt basin and that resulting shortening in the distal part is partly accommodated by horizontal compaction. Models with moderately high fluid pressure best reproduce conditions considered to have been necessary for large‐scale gravitational spreading in the northwestern Gulf of Mexico, which led to the formation of the Perdido Fold Belt.

Effective stress and pore pressure in the Nankai accretionary prism off the Muroto Peninsula, southwestern Japan

We developed a theoretical method for predicting effective stress and pore pressure based on rock physics model. We applied the method to reveal the pore pressure distribution within the Nankai accretionary prism off southwestern Japan and to investigate variations in pore pressure associated with evolution of the plate boundary décollement. From the crack aspect ratio spectrum estimated from laboratory and well‐log data, we calculated a theoretical relationship between acoustic velocity and mean effective stress by using differential effective medium theory. By iteratively fitting the theoretically calculated velocity to the seismic velocities derived from 3D tomographic inversion, we estimated in situ mean effective stress within the accretionary prism. Pore pressure is then the difference between the effective stress and the confining stress. When we calculated pore pressure, we considered compressive state of stress in the accretionary prism. Our results confirm that pore fluid pressure is high within the subducting sedimentary sequence below the décollement; we determined a normalized pore pressure ratio (λ*) of 0.4–0.7. Abnormal pore pressures develop in the underthrust sequence as a result of the increase in overburden load because of the thickened overlying prism and a low permeability barrier across the décollement. Overpressuring within the accreted sequence is initiated at the deformation front and proceeds landward. The increase in horizontal compaction within the accreted sequence may raise pore pressures within the accreted sequence, and the pore pressure (mean effective stress) contrast at the décollement becomes smaller landward of the deformation front.

Geomechanics of polygonal fault systems: a review

ABSTRACT Layer-bound systems of polygonal faults are found in sequences of very fine-grained sediments that have typically undergone passive subsidence and burial. In the absence of tectonic extension, the heave of the faults must be complemented by horizontal compaction of the sediments. Density inversion, syneresis and low coefficients of friction on fault planes have all been proposed as causal mechanisms for the development of polygonal fault systems, but most sequences that contain polygonal faults are not underlain by sediments of lower density and there is a lack of evidence to support the idea that syneresis is responsible. Laboratory measurements of clay properties and a recent field test based on well data strongly suggest that low coefficients of residual friction in fine-grained sediments are key to the growth of faults that eventually develop into polygonal systems. However, coefficients of residual friction apply to faults only after initial slip has taken place, so some other mechanism must be responsible for the initial nucleation of the faults. Various speculative suggestions have been made, but there is no evidence that nucleation of those faults that evolve into polygonal systems differs fundamentally from the processes involved in the nucleation of other faults in soft sediments.

Relief and shortening in detachment folds

We present new thickness-relief methods for determining shortening S( z), mean shortening S ¯ ( z ) and curvimetric shortening S c( z) as a function of height in well-imaged structures. Thickness-relief measurements allow us to constrain the fractions of the deformation that are layer-parallel pure shear, simple shear, and flexural, and to measure the excess area A e of fold cores caused, for example, by flow of evaporites. The key measurements are made in the thickness domain, which allows for more certain determination of regional stratigraphic gradients, which is required for reliable separation of structural relief from primary stratigraphic thickness variations. We apply these methods to a diverse set of active detachment folds from the fronts of the Nankai trough Japan, Cascadia accretionary wedge Oregon, southern Tianshan China and Agbami anticline deep-water Niger delta. Three of these folds can be approximated by pure-shear detachment fold models because more than 95% of the shortening is by layer-parallel heterogeneous pure shear plus horizontal compaction. For this reason curvimetric shortening is one to two orders of magnitude less than total shortening. In contrast, Agbami anticline shows layer-parallel stretching above the excess area of the fold core. In all examples the layer-parallel simple-shear component is negligible or absent.

Effects of horizontal strain in estimating specific storage and compaction in confined and leaky aquifer systems

Mathematical and numerical distinctions are made between developments of governing equations involving groundwater flow and granular displacement where (1) only vertical strain and no change in total normal load are assumed, and (2) such limitations are eliminated. The former is referred to here as the Jacob-Terzaghi method; the latter is referred to as the Biot method. Numerical comparisons of drawdown, compaction, and strain are made for hypothetical confined and leaky aquifer systems for one- and three-dimensional strain problems. Simulation results indicate that incorporation of horizontal strain results in reduced drawdowns. Vertical compaction for a specified stress distribution is greatly reduced when horizontal strain is invoked. The distribution of strain components indicates that approximately two-thirds of the total volume strain originates from horizontal compaction in pumped aquifers. Incorporation of horizontal strain also affects the distribution of flux through the overlying confining layer as a function of distance from the pumping well and may ultimately affect the interpretation of aquifer-test data.

Hydraulic Fracturing At Sedimentary Basin Scale

One key point for simulating the hydraulic fracturing at basin scale, is to be able to compute the stress tensor. This is generally not addressed in basin model because of the complexity of this problem. In order to get access to the stress tensor we have to assume that:- one of the principal stress is vertical and equals the overburden weight;- the horizontal stress is deduced from the vertical stress with the K0 coefficient that is a function of depth and of the tectonical setting. Consolidation is considered here as the combined effect of the mechanical compaction and the chemical compaction. The mechanical compaction is mainly caused by the rearrangement of grains during burial and could be represented at the macroscopical scale by an elastoplastic rheology. The chemical compaction is considered here as resulting from dissolution-precipitation mechanisms, generally induced by stress (pressure-solution). The chemical compaction could be represented at the macroscopical scale by a viscoplastic rheology. The complete elastoplastic yield is defined by the union of the consolidation elastoplastic yield and of the different failure criteria that could be seen as elastobrittle yields. Thus, the elastoplastic yield is composed of six elementary elastoplastic yields which define the onset of vertical compaction, horizontal compaction, vertical tensile fracturing, horizontal tensile fracturing, subvertical shear fracturing, and subhorizontal shear fracturing. Due to the consolidation, most of the parameters that describe the physical properties of the sediments evolve with the geological times. One difficulty is to quantify the degree of evolution of the porous medium during its geological history. Here, we have chosen to measure the evolution of the sediments by their porosity. The local simulations showed that fracturing may occur is numerous configurations. Some of these configurations indicate that the sediments can reach the limit of its elastic domain and then may initiate some fracturing. In other cases, the sediment changes its loading path but stay at the elastoplastic yield surface. In this case the theoretical considerations are not valid anymore and fracture criteria should be defined in the plastic domain.

A boltzmann machine approach to code optimization

In this paper we present a statistical approach to the problem of horizontal code compaction for the class of parallel synchronous non-homogeneous architectures. The proposed technique is based on a formulation of the problem using the Boltzmann Machine model. The formalization of the compaction process in terms of a non-deterministic connectionist framework enables to overcome the NP-hardness of the problem and to avoid the introduction of specialized heuristics which necessarily limit the generality of other techniques. Compilations of some Livermore loop kernels are presented, showing how the quality of the code produced with a Boltzmann Machine is comparable with a hand-compacted one. These results prove the skills of the method in implementing an effective alternative to traditional compilation techniques.