Simulation Library Research Articles

Digital Sustainability of Intangible Cultural Heritage: The Example of the “Wu Leno” Weaving Technique in Suzhou, China

Digital technology can be used to enable the living preservation and heritage innovation of the intangible cultural heritage of Chinese traditional textile skills for sustainable development. Taking the “Wu Leno” weaving technique in Suzhou as the research object, we first elaborated on the origin, development, and evolution of “Wu Leno” through documentary research and case study methods. Second, considering the inheritance of the Leno weaving technique, we used digital technology to establish the pattern. The study showed that a digital database and virtual reality technology can be used to build a loom model and create a simulation library to realize innovative applications and interactive experiences in apparel design. The study also showed that the digital database and the interactive technology of virtual reality can provide practical pathways and explore experiences for the preservation, inheritance, cross-border communication, and sustainable innovation development of Suzhou “Wu Leno” weaving techniques in modern times.

Design and Construction of Electronic Muscle Stimulator (EMS)

The project is titled Design and Construction of Electronic Muscle Stimulator (EMS) with embedded time Display unit and Dual Source of Powering. The EMS is a type of electrotherapy device that functions by stimulating muscle contraction using electrical impulses to strengthen weak muscles, reduce swelling, aid in blood circulations that helps to heal wounds. These impulses are generated by the device and deliver through electrode pads over the area that requires stimulation. The impulses from EMS mimic the action potential (the stimulus required to contract the muscle) that naturally comes from central nervous system (CNS). Proteus, software simulation Library was used to simulate the components and to understand the behaviour of the output stage. After that, hardware components were assembled together that gave rise to this EMS apparatus. Designing of the device was completed with different modes tagged Press, Thump, Rub, and vibrate with frequencies 40Hz, 20Hz, 10Hz, and 3Hz respectively. The project was completed and human experiment was carried and result showed that decrease in impedances brings increase in current intensity.

Numerical modelling and evaluation of a novel sorption module for thermally driven heat pumps and chillers using open-source simulation library

Sorption modules are the core component of thermally driven heat pumps and chillers. The efficiency of these devices strongly depends on the advantageous design of sorption modules. In this paper a calibrated and validated numerical model for an innovative sorption module with a combined evaporator-condenser is presented. The adsorption heat exchanger is based on flat tube-lamella design directly crystallized with the zeotype adsorbent silico-alumino-phosphate-34. The prediction quality of the model regarding the efficiency is within the measurement uncertainty (±0.02 kJ/kJ). Besides the good prediction quality of this integral performance indicator, the root mean square deviation of the transient outlet temperatures is in the range of 1.1…1.9 K, which is a very good agreement. Since the performance indicators efficiency and power density strongly depend on the temperature boundary conditions and half cycle times, an in-depth analysis of the experimental data using the method of heat and mass transfer resistances is suggested that overcomes this limitation. This analysis allows for a direct comparison with other sorption module designs. In a first step this analysis revealed that the evaporator-condenser component limits the sorption process during evaporation. Compared to other designs the evaporator-condenser has a 3–5 times higher volume scaled heat and mass transfer resistance (17 dm3K/kW) in the evaporation phase underlining the necessity to further optimize this component in future modules.

UPAFuzzySystems: A Python Library for Control and Simulation with Fuzzy Inference Systems

The main goal of control theory is input tracking or system stabilization. Different feedback-computed controlled systems exist in this area, from deterministic to soft methods. Some examples of deterministic methods are Proportional (P), Proportional Integral (PI), Proportional Derivative (PD), Proportional Integral Derivative (PID), Linear Quadratic (LQ), Linear Quadratic Gaussian (LQG), State Feedback (SF), Adaptative Regulators, and others. Alternatively, Fuzzy Inference Systems (FISs) are soft-computing methods that allow using the human expertise in logic in IF–THEN rules. The fuzzy controllers map the experience of an expert in controlling the plant. Moreover, the literature shows that optimization algorithms allow the adaptation of FISs to control different processes as a black-box problem. Python is the most used programming language, which has seen the most significant growth in recent years. Using open-source libraries in Python offers numerous advantages in software development, including saving time and resources. In this paper, we describe our proposed UPAFuzzySystems library, developed as an FISs library for Python, which allows the design and implementation of fuzzy controllers with transfer-function and state-space simulations. Additionally, we show the use of the library for controlling the position of a DC motor with Mamdani, FLS, Takagi–Sugeno, fuzzy P, fuzzy PD, and fuzzy PD-I controllers.

Vespa: Integrated applications for RF pulse design, spectral simulation and MRS data analysis.

The Vespa package (Versatile Simulation, Pulses, and Analysis) is described and demonstrated. It provides workflows for developing and optimizing linear combination modeling (LCM) fitting for 1 H MRS data using intuitive graphical user interface interfaces for RF pulse design, spectral simulation, and MRS data analysis. Command line interfaces for embedding workflows in MR manufacturer platforms and utilities for synthetic dataset creation are included. Complete provenance is maintained for all steps in workflows. Vespa is written in Python for compatibility across operating systems. It embeds the PyGAMMA spectral simulation library for spectral simulation. Multiprocessing methods accelerate processing and visualization. Applications use the Vespa database for results storage and cross-application access. Three projects demonstrate pulse, sequence, simulation, and data analysis workflows: (1) short TE semi-LASER single-voxel spectroscopy (SVS)LCM fitting, (2) optimizing MEGA-PRESS (MEscher-GArwood Point RESolved Spectroscopy) flip angle and LCM fitting, and (3) creating a synthetic short TE dataset. The LCM workflows for in vivo basis set creation and spectral analysis showed reasonable results for both the short TE semi-LASER and MEGA-PRESS. Examples of pulses, simulations, and data fitting are shown in Vespa application interfaces for various steps to demonstrate the interactive workflow. Vespa provides an efficient and extensible platform for characterizing RF pulses, pulse design, spectral simulation optimization, and automated LCM fitting via an interactive platform. Modular design and command line interface make it easy to embed in other platforms. As open source, it is free to the MRS community for use and extension. Vespa source code and documentation are available through GitHub.

Towards auto-tuning Multi-Site Molecular Dynamics simulations with AutoPas

There exists an extensive literature of algorithms for short-range pairwise interactions in particle simulations, however, there is no single algorithm that performs the most optimally in every scenario, motivating the use of auto-tuning to select the optimal pairwise interaction algorithm. Previous efforts to auto-tune Molecular Dynamics have focused on Single-Site Molecular Dynamics, where the computational cost for the intermolecular force calculation is constant. Alternatively, for Multi-Site Molecular Dynamics, the cost of this calculation varies with the number of sites, which, as we show in this paper, can result in different optimal algorithms. Despite this further benefit for auto-tuning, it has yet to be applied to Multi-Site Molecules. In this paper, we introduce an implementation of Multi-Site Molecular Dynamics that is integrated with AutoPas. Using this implementation, we analyse how the relative performance between these algorithms varies as the number of sites varies, for both homogeneous and heterogeneous molecule distributions, and for two different hardware. Furthermore, we demonstrate the advantage of auto-tuning in the context of Multi-Site Molecular Dynamics using the node-level short-range particle simulation library, AutoPas.

Library of Simulated Gamma‐Ray Glows and Application to Previous Airborne Observations

AbstractGamma‐Ray Glows (GRGs) are high energy radiation originating from thunderclouds, in the MeV energy regime, with typical duration of seconds to minutes, and sources extended over several to tens of square kilometers. GRGs have been observed from detectors placed on ground, inside aircraft and on balloons. In this paper, we present a general purpose Monte‐Carlo model of GRG production and propagation. This model is first compared to a model from Zhou et al. (2016, https://doi.org/10.1016/j.astropartphys.2016.08.004) relying on another Monte‐Carlo framework, and small differences are observed. We then have built an extensive simulation library, made available to the community. This library is used to reproduce five previous gamma‐ray glow observations, from five airborne campaigns: balloons from Eack et al. (1996b, https://doi.org/10.1029/96gl02570), Eack et al. (2000, https://doi.org/10.1029/1999gl010849); and aircrafts from ADELE (Kelley et al., 2015, https://doi.org/10.1038/ncomms8845), ILDAS (Kochkin et al., 2017, https://doi.org/10.1002/2017jd027405) and ALOFT (Østgaard et al., 2019, https://doi.org/10.1029/2019jd030312). Our simulation results confirm that fluxes of cosmic‐ray secondary particles present in the background at a given altitude can be enhanced by several percent (MOS process), and up to several orders of magnitude (RREA process) due to the effect of thunderstorms' electric fields, and explain the five observations. While some GRG could be explained purely by the MOS process, E‐fields significantly larger than Eth are required to explain the strongest GRGs observed. Some of the observations also came with in‐situ electric field measurements, that were always lower than Eth, but may not have been obtained from regions where the glows are produced. This study supports the claim that kilometer‐scale E‐fields magnitudes of at least the level of Eth must be present inside some thunderstorms.

OpenDust: A fast GPU-accelerated code for the calculation of forces acting on microparticles in a plasma flow

We present the first open-source, GPU-based code for complex plasmas. The code, OpenDust, pursues to provide researchers, both experimenters and theorists, a user-friendly and high-performance tool for self-consistent calculation forces, acting on microparticles, and microparticles' charges in a plasma flow. OpenDust performance originates from a highly-optimized Cuda back-end and performs self-consistent calculations of plasma flow around microparticles in seconds. This code outperforms all available codes for self-consistent complex plasma simulation. Moreover, OpenDust can also be used for the simulation of larger systems of dust microparticles, which was previously unavailable. OpenDust interface is written in Python, which provides ease-of-use and simple installation from the Conda repository. Program summaryProgram Title: OpenDustCPC Library link to program files:https://doi.org/10.17632/bs7rthk29w.1Developer's repository link:https://github.com/kolotinsky1998/opendustCode Ocean capsule:https://codeocean.com/capsule/2557151Licensing provisions: MITProgramming language: PythonNature of problem: GPU cards can significantly speedup self-consistent calculations of forces, acting on microparticles in a plasma flow. However, the available codes are CPU-based, or not provided as a package that can also be easily used. Therefore, researchers need to spend much time writing their own codes or use less effective ones.Solution method: Development of a highly-optimized GPU-accelerated library for self-consistent simulations of streaming plasma around microparticles. The functionality of the library is available through a Python interface, which makes it easy to use.

APLUS: A Python library for usefulness simulations of machine learning models in healthcare

Despite the creation of thousands of machine learning (ML) models, the promise of improving patient care with ML remains largely unrealized. Adoption into clinical practice is lagging, in large part due to disconnects between how ML practitioners evaluate models and what is required for their successful integration into care delivery. Models are just one component of care delivery workflows whose constraints determine clinicians’ abilities to act on models’ outputs. However, methods to evaluate the usefulness of models in the context of their corresponding workflows are currently limited. To bridge this gap we developed APLUS, a reusable framework for quantitatively assessing via simulation the utility gained from integrating a model into a clinical workflow. We describe the APLUS simulation engine and workflow specification language, and apply it to evaluate a novel ML-based screening pathway for detecting peripheral artery disease at Stanford Health Care.

Multi-Method Simulation and Multi-Objective Optimization for Energy-Flexibility-Potential Assessment of Food-Production Process Cooling

Process cooling for food production is an energy-intensive industry with complex interactions and restrictions that complicate the ability to utilize energy-flexibility due to unforeseen consequences in production. Therefore, methods for assessing the potential flexibility in individual facilities to enable the active participation of process-cooling facilities in the electricity system are essential, but not yet well discussed in the literature. Therefore, this paper introduces an assessment method based on multi-method simulation and multi-objective optimization for investigating energy flexibility in process cooling, with a case study of a Danish process-cooling facility for canned-meat food production. Multi-method simulation is used in this paper: multi-agent-based simulation to investigate individual entities within the process-cooling system and the system’s behavior; discrete-event simulation to explore the entire process-cooling flow; and system dynamics to capture the thermophysical properties of the refrigeration unit and states of the refrigerated environment. A simulation library is developed, and is able to represent a generic production-flow of the canned-food process cooling. A data-driven symbolic-regression approach determines the complex logic of individual agents. Using a binary tuple-matrix for refrigeration-schedule optimization, the refrigeration-cycle operation is determined, based on weather forecasts, electricity price, and electricity CO2 emissions without violating individual room-temperature limits. The simulation results of one-week’s production in October 2020 show that 32% of energy costs can be saved and 822 kg of CO2 emissions can be reduced. The results thereby show the energy-flexibility potential in the process-cooling facilities, with the benefit of overall production cost and CO2 emissions reduction; at the same time, the production quality and throughput are not influenced.

Evaluation of Portable Programming Models to Accelerate LArTPC Detector Simulations

The Liquid Argon Time Projection Chamber (LArTPC) technology is widely used in high energy physics experiments, including the upcoming Deep Underground Neutrino Experiment (DUNE). Accurately simulating LArTPC detector responses is essential for analysis algorithm development and physics model interpretations. Accurate LArTPC detector response simulations are computationally demanding, and can become a bottleneck in the analysis workflow. Compute devices such as General-Purpose Graphics Processing Units (GPGPUs) have the potential to substantially accelerate simulations compared to traditional CPU-only processing. The software development for these compute accelerators often carries the cost of specialized code refactorization and porting to match the target hardware architecture. With the rapid evolution and increased diversity of the computer architecture landscape, it is highly desirable to have a portable solution that also maintains reasonable performance. We report our ongoing effort in evaluating Kokkos as a basis for this portable programming model using LArTPC simulations in the context of the Wire-Cell Toolkit, a C++ library for LArTPC simulations, data analysis, reconstruction and visualization.

Tidal Disruption Events from Eccentric Orbits and Lessons Learned from the Noteworthy ASASSN-14ko

Stars grazing supermassive black holes (SMBHs) on bound orbits may survive tidal disruption, causing periodic flares. Inspired by the recent discovery of the periodic nuclear transient ASASSN-14ko, a promising candidate for a repeating tidal disruption event (TDE), we study the tidal deformation of stars approaching SMBHs on eccentric orbits. With both analytical and hydrodynamic methods, we show the overall tidal deformation of a star is similar to that in a parabolic orbit provided that the eccentricity is above a critical value. This allows one to make use of existing simulation libraries from parabolic encounters to calculate the mass fallback rate in eccentric TDEs. We find the flare structures of eccentric TDEs show a complicated dependence on both the SMBH mass and the orbital period. For stars orbiting SMBHs with relatively short periods, we predict significantly shorter-lived duration flares than those in parabolic TDEs, which can be used to predict repeating events if the mass of the SMBH can be independently measured. Using an adiabatic mass-loss model, we study the flare evolution over multiple passages, and show the evolved stars can survive many more passages than main-sequence stars. We apply this theoretical framework to the repeating TDE candidate ASASSN-14ko and suggest that its recurrent flares originate from a moderately massive (M ≳ 1 M ⊙), extended (likely ≈10 R ⊙), evolved star on a grazing, bound orbit around the SMBH. Future hydrodynamic simulations of multiple tidal interactions will enable realistic models on the individual flare structure and the evolution over multiple flares.

Characterizing Experimental Monoclonal Antibody Interactions and Clustering Using a Coarse-Grained Simulation Library and a Viscosity Model.

Attractive protein-protein interactions in concentrated monoclonal antibody (mAb) solutions may lead to the formation of clusters that increase viscosity. Here, we propose an analytical model that relates mAb solution viscosity to clustering by accounting for the contributions of suboptimal mAb packing within a cluster and cluster fractal dimension. The influence of short-range, anisotropic attractions and long-range Coulombic repulsion on cluster properties is investigated by analyzing the cluster-size distributions, cluster fractal dimensions, radial distribution functions, and static structure factors from a library of coarse-grained molecular dynamics simulations. The library spans a vast range of mAb charges and attractive interactions in solutions of varying ionic strength. We present a framework for combining the viscosity model and simulation library to successfully characterize the attraction, repulsion, and clustering of an experimental mAb in three different pH and cosolute conditions by fitting the measured viscosity or structure factor from small-angle X-ray scattering. At low ionic strength, the cluster-size distribution is impacted by strong charges, and both the viscosity and net charge or structure factor and net charge must be considered to deconvolute the effects of short-range attraction and long-range repulsion.

SERIAL AND DIFFERENT PARALLEL IMPLEMENTATIONS OF LATTICE BOLTZMANN METHOD IN PYTHON: A COMPARATIVE ANALYSIS

The lattice Boltzmann method (LBM) is known for its capability to model complex fluid systems. Python is becoming one of the most valuable tools for scientific programming and prototyping of commercial software. In parallel, there is nearly no information on LBM simulation libraries in Python. The current study addresses the issue of Pythonbased LBM simulations. We have developed and tested the serial and two concurrent approaches (multiprocessing and multithreading) through 7 different versions of the LBM solvers for a range of the various numbers of nodes. Parallelization is also defined for three individual levels of domain decomposition, and RunTime profile data for entire solvers have been discussed comparatively in detail. Apart from the implementation strategy, a preliminary comparison is made between two data storage styles to investigate the effect of data shape on RunTime. In contrast with the literature, serial processing provides the fastest results for task execution of LBM simulation in Python due to the strengths of Python in data manipulation. One version of multithreaded solver also provides the same results with serial implementation; however, in conclusion, and due to the ease of serial programming, serial implementation of LBM in Python is defined as the most practical candidate.

LibRoadRunner 2.0: a high performance SBML simulation and analysis library.

This article presents libRoadRunner 2.0, an extensible, high-performance, cross-platform, open-source software library for the simulation and analysis of models expressed using the systems biology markup language (SBML). libRoadRunner is a self-contained library, able to run either as a component inside other tools via its C++, C and Python APIs, or interactively through its Python or Julia interface. libRoadRunner uses a custom just-in-time (JIT) compiler built on the widely used LLVM JIT compiler framework. It compiles SBML-specified models directly into native machine code for a large variety of processors, making it fast enough to simulate extremely large models or repeated runs in reasonable timeframes. libRoadRunner is flexible, supporting the bulk of the SBML specification (except for delay and non-linear algebraic equations) as well as several SBML extensions such as hierarchical composition and probability distributions. It offers multiple deterministic and stochastic integrators, as well as tools for steady-state, sensitivity, stability and structural analyses. libRoadRunner binary distributions for Windows, Mac OS and Linux, Julia and Python bindings, source code and documentation are all available at https://github.com/sys-bio/roadrunner, and Python bindings are also available via pip. The source code can be compiled for the supported systems as well as in principle any system supported by LLVM-13, such as ARM-based computers like the Raspberry Pi. The library is licensed under the Apache License Version 2.0.

PDynamo3 Molecular Modeling and Simulation Program.

pDynamo3 is the first formal version of the Dynamo molecular modeling and simulation library that is written in Python 3. It follows from the previous pDynamo versions written in Python 2, the first of which was released in 2007. Both pDynamo and its predecessor, fDynamo, were designed with the aim of providing easy-to-use and flexible frameworks for performing molecular simulations at an atomistic level with a special emphasis on those employing hybrid quantum chemical and molecular mechanical potential methods. Although the use of pDynamo3 is quite similar to that of pDynamo2, it has added significant new capability and also undergone extensive restructuring that will make it much easier to extend with new functionality. The pDynamo3 code is issued under the GNU general public license at https://github.com/pdynamo/pDynamo3 with additional information on the pDynamo website https:www.pdynamo.org.

Forward-Solution Noninvasive Computational Arrhythmia Mapping: The VMAP Study.

The accuracy of noninvasive arrhythmia source localization using a forward-solution computational mapping system has not yet been evaluated in blinded, multicenter analysis. This study tested the hypothesis that a computational mapping system incorporating a comprehensive arrhythmia simulation library would provide accurate localization of the site-of-origin for atrial and ventricular arrhythmias and pacing using 12-lead ECG data when compared with the gold standard of invasive electrophysiology study and ablation. The VMAP study (Vectorcardiographic Mapping of Arrhythmogenic Probability) was a blinded, multicenter evaluation with final data analysis performed by an independent core laboratory. Eligible episodes included atrial and ventricular: tachycardia, fibrillation, pacing, premature atrial and ventricular complexes, and orthodromic atrioventricular reentrant tachycardia. Mapping system results were compared with the gold standard site of successful ablation or pacing during electrophysiology study and ablation. Mapping time was assessed from time-stamped logs. Prespecified performance goals were used for statistical comparisons. A total of 255 episodes from 225 patients were enrolled from 4 centers. Regional accuracy for ventricular tachycardia and premature ventricular complexes in patients without significant structural heart disease (n=75, primary end point) was 98.7% (95% CI, 96.0%-100%; P<0.001 to reject predefined H0 <0.80). Regional accuracy for all episodes (secondary end point 1) was 96.9% (95% CI, 94.7%-99.0%; P<0.001 to reject predefined H0 <0.75). Accuracy for the exact or neighboring segment for all episodes (secondary end point 2) was 97.3% (95% CI, 95.2%-99.3%; P<0.001 to reject predefined H0 <0.70). Median spatial accuracy was 15 mm (n=255, interquartile range, 7-25 mm). The mapping process was completed in a median of 0.8 minutes (interquartile range, 0.4-1.4 minutes). Computational ECG mapping using a forward-solution approach exceeded prespecified accuracy goals for arrhythmia and pacing localization. Spatial accuracy analysis demonstrated clinically actionable results. This rapid, noninvasive mapping technology may facilitate catheter-based and noninvasive targeted arrhythmia therapies. URL: https://www. gov; Unique identifier: NCT04559061.

Dynamic simulation of the first cool down of the JT-60SA cryo-magnet system

In the framework of the Broader Approach research and development for fusion activities, involving Japan and Europe, the JT-60SA tokamak has been jointly constructed in Naka, Japan to support the operation of ITER and to study advanced plasma scenarios for future fusion power plants. The integrated commissioning of the tokamak started mid 2020 with a Japanese and European team, composed of QST, F4E and Eurofusion. CEA, as member of Eurofusion consortium, has brought expertise and support in the operation of the cryo-magnetic system.The cool-down from 300 K to 4.5 K took more than one month following stringent temperature constraints on the magnets and the thermal shields. Dynamic simulations models with Simcryogenics were developed during the cool down to support the magnet cool-down operation. Simcryogenics is the simulation library developed by CEA (Department of Low Temperature Systems) on a Matlab/Simulink platform for modelling cryogenic processes. This article presents the first cool-down of the JT-60SA magnet system and attempts to reproduce the measured data from 300 K to 4.5 K with the Simcryogenics models.

The implementation of discrete-event simulation and demand forecasting using Temporal Fusion Transformers to validate spare parts inventory policy for petrochemicals industry

One of the important strategic factors of petrochemicals plant maintenance is the spare parts inventory policy, It is significant for the efficiency, reliability, and productivity of the petrochemicals industry. An unsuitable spare part inventory policy will lead to a loss in long engineering machinery downtime due to a shortage of spare parts. To implement the spare parts inventory policy which is able to fulfill the future demand for spare parts, calculation by various statistical theories and working processes is used to custom the spare parts inventory policy. However, to validate whether or not the customized spare parts inventory policy is suitable, the discrete-event simulation library SimPy is used to mimic the actual spare parts inventory system. It must be involved in the performance evaluation process of the customized spare parts inventory policy. The inventory simulation model consists of many events depending on the supply chain system. The crucial event which is the most complex for the simulation of spare parts inventory is the demand event. This work applies the demand forecasting technique to the simulation by using deep learning with a prebuilt architecture model called Temporal Fusion Transformers (TFT). The averaged MAE of the point predictions from a global model is 0.4874+/- 6.7744 on the validation dataset and 0.6424+/-3.4963 on the test dataset. Our method predicts a quantile forecast of the future demand which is able to handle the stochastic nature of the spare parts demand in the petrochemicals industry and the result from the simulation outcome is more accurate and close to the outcome from an actual inventory system. The information from the analysis of the simulation outcome is used by the inventory management team to make decisions about the custom inventory policy before deploying it to the actual system.

The implementation of discrete-event simulation and demand forecasting using Temporal Fusion Transformers to validate spare parts inventory policy for petrochemicals industry

One of the important strategic factors of petrochemicals plant maintenance is the spare parts inventory policy, It is significant for the efficiency, reliability, and productivity of the petrochemicals industry. An unsuitable spare part inventory policy will lead to a loss in long engineering machinery downtime due to a shortage of spare parts. To implement the spare parts inventory policy which is able to fulfill the future demand for spare parts, calculation by various statistical theories and working processes is used to custom the spare parts inventory policy. However, to validate whether or not the customized spare parts inventory policy is suitable, the discrete-event simulation library SimPy is used to mimic the actual spare parts inventory system. It must be involved in the performance evaluation process of the customized spare parts inventory policy. The inventory simulation model consists of many events depending on the supply chain system. The crucial event which is the most complex for the simulation of spare parts inventory is the demand event. This work applies the demand forecasting technique to the simulation by using deep learning with a prebuilt architecture model called Temporal Fusion Transformers (TFT). The averaged MAE of the point predictions from a global model is 0.4874+/- 6.7744 on the validation dataset and 0.6424+/-3.4963 on the test dataset. Our method predicts a quantile forecast of the future demand which is able to handle the stochastic nature of the spare parts demand in the petrochemicals industry and the result from the simulation outcome is more accurate and close to the outcome from an actual inventory system. The information from the analysis of the simulation outcome is used by the inventory management team to make decisions about the custom inventory policy before deploying it to the actual system.