Small-scale Sets Research Articles

Modified rapid small-scale fractionation method for quantitatively separating pyrrolic nitrogen compounds, phenols-ketones and carboxylic acids fractions from petroleum

This research outlines a fast and efficient technique for separating the pyrrolic, phenolic-ketonic, and carboxylic acid fractions from oils in a small-scale setting. This approach enables the separation and analysis of neutral nitrogen fractions that contain indoles, carbazoles, and benzocarbazoles, as well as ketonic-phenolic fractions that contain phenols and ketones. The study showcases the potential for accurately measuring petroleum composition. The reliability of the fractions’ concentration results is determined by the mass of the separated fraction, with completely reliable results achievable for masses exceeding 1.4 mg. The fraction’s mass can also be used to evaluate the relative expanded uncertainty of a single separation procedure. As the mass and concentration of the fraction increase, the relative expanded uncertainty of the concentration decreases, and this can be modelled using a power function.

A Self-Training-Based System for Die Defect Classification

With increasing wafer sizes and diversifying die patterns, automated optical inspection (AOI) is progressively replacing traditional visual inspection (VI) for wafer defect detection. Yet, the defect classification efficacy of current AOI systems in our case company is not optimal. This limitation is due to the algorithms’ reliance on expertly designed features, reducing adaptability across various product models. Additionally, the limited time available for operators to annotate defect samples restricts learning potential. Our study introduces a novel hybrid self-training algorithm, leveraging semi-supervised learning that integrates pseudo-labeling, noisy student, curriculum labeling, and the Taguchi method. This approach enables classifiers to autonomously integrate information from unlabeled data, bypassing the need for feature extraction, even with scarcely labeled data. Our experiments on a small-scale set show that with 25% and 50% labeled data, the method achieves over 92% accuracy. Remarkably, with only 10% labeled data, our hybrid method surpasses the supervised DenseNet classifier by over 20%, achieving more than 82% accuracy. On a large-scale set, the hybrid method consistently outperforms other approaches, achieving up to 88.75%, 86.31%, and 83.61% accuracy with 50%, 25%, and 10% labeled data. Further experiments confirm our method’s consistent superiority, highlighting its potential for high classification accuracy in limited-data scenarios.

A techno-economic analysis of future hydrogen reconversion technologies

The transformation of fossil fuel-based power generation systems towards greenhouse gas-neutral ones based on renewable energy sources is one of the key challenges facing contemporary society. The temporal volatility that accompanies the integration of renewable energy (e.g. solar radiation and wind) must be compensated to ensure that at any given time, a sufficient supply of electrical energy for the demands of different sectors is available. Green hydrogen, which is produced using renewable energy sources via electrolysis, can be used to chemically store electrical energy on a seasonal basis. Reconversion technologies are needed to generate electricity from stored hydrogen during periods of low renewable electricity generation. This study presents a detailed techno-economic assessment of hydrogen gas turbines. These technologies are also superior to fuel cells due to their comparatively low investment costs, especially when it comes to covering the residual loads. As of today, hydrogen gas turbines are only available in laboratory or small-scale settings and have no market penetration or high technology readiness level. The primary focus of this study is to analyze the effects on gas turbine component costs when hydrogen is used instead of natural gas. Based on these findings, an economic analysis addressing the current state of these turbine components is conducted. A literature review on the subsystems is performed, considering statements from leading manufactures and researchers to derive the cost deviations and total cost per installed capacity (€/kWel). The results reveal that a hydrogen gas turbine power plant has an expected cost increase of 8.5% compared to a conventional gas turbine one. This leads to an average cost of 542.5 €/kWel for hydrogen gas turbines. For hydrogen combined cycle power plants, the expected cost increase corresponds to the cost of the gas turbine system, as the steam turbine subsystem remains unaffected by fuel switching. Additionally, power plant retrofit potentials were calculated and the respective costs in the case of an upgrade were estimated. For Germany as a case study for an industrialized country, the potential of a possible retrofit is between 2.7 and 11.4 GW resulting to a total investment between 0.3 and 1.1 billion €.

소규모 초등학교 학생의 방과후학교 스포츠 활동 경험과 교육적 의미

Objectives The purpose of this study was to examine the after-school sports activity experiences of elementary school students in a small-scale setting and explore the educational implications of these experiences. Methods To achieve this, data were collected from 6 students in grades 3 to 6 who participated in after-school sports activities at a small-scale elementary school during the first semester from March to July 2023. Data col-lection involved participant observation, in-depth interviews, and document analysis. The data were analyzed us-ing iterative comparative analysis, and the credibility of the research was ensured through triangulation, peer dis-cussions, and confirmation from research participants. Results The students' experience in small-scale elementary school after-school sports activities is described as “Reviving the Spark of Small-Scale School Physical Education.” Due to insufficient preparation in regular physical education classes by homeroom teachers, students had lost interest. In response, after-school sports activities served the purpose of fulfilling the lack of physical activity, bridging care gaps, and providing enjoyable leisure time, positively impacting students' holistic development and self-realization. The educational significance is framed as “Illuminating the Torch of Public Education.” The collaborative format of after-school sports activities facilitated interaction between grade levels, fostering exchange among students. It provided opportunities for mutual learning and motivation between juniors and seniors. Additionally, it contributed to reducing private educa-tion expenses, restoring trust in public education, and offering a broad range of sports experiences. Conclusions The after-school sports activities in small-scale elementary schools provide students with oppor-tunities for growth and development on various levels. Therefore, the activation of physical education activities in small-scale elementary schools is necessary.

Improving spatial transferability of deep learning models for small-field crop yield prediction

Predicting crop yield using deep learning (DL) and remote sensing is a promising technique in agriculture. In smallholder agriculture (<2 ha), where 84% of the farms operate globally, it is crucial to build a model that can be useful across several fields (high spatial transferability). However, enhancing spatial model transferability in a small-scale setting faces significant challenges, including spatial autocorrelation, heterogeneity and scale dependence of spatial dynamics, as well as the need to address limited data points. This study aimed to test the hypothesis that spatial cross validation (SCV) is a more suitable model validation practice than random cross validation (RCV) to enhance model transferability for spatial prediction in a small-scale farming setting. We compared the performances of DL models that predict crop yield for several settings including three crop types and two DL architectures based on RCV with and without overlapping samples and SCV. Notably, we conducted model performance tests on external, equally sized fields instead of the field used for training. We used high resolution RGB imagery taken with a drone as input. Our results show that the models using SCV outperformed those using RCV when the models were tested on external fields (on average r = 0.37 for SCV, r = 0.18 for RCV with overlap and r = 0.07 without), even though the models using SCV showed a substantially lower performance for cross validation (CV) than those using RCV (r with SCV and RCV w/o overlap = 0.73 and 0.98/0.73, respectively). The results suggest that RCV leads to over-optimism by overfitting the spatial structure and remembering image-specific information (so called memorization). Our study offers the first empirical evidence in agriculture that SCV is preferable to RCV in small field settings for making DL models more transferable.

P182: Effects of Physical Environment on Quality of Life of Residents in Dementia Facilities in Canada &amp; South Korea: A Longitudinal Observational Study

Approximately 44% of new residents of care facilities in Korea were diagnosed with dementia (Song, Park &amp; Kim, 2013), and in Canada, about one-third of older adults younger than 80 who have been diagnosed with dementia live in long-term care facilities (Canadian Institute for Health Information, 2018). Due to the rapid increase of these figures in the future, continuing to provide assistance services and appropriate environment for residents with dementia could be challenging for both countries.This longitudinal observational study aims to examine whether residents with dementia in long-term care facilities with variability in physical environment attributions in Vancouver (N=11), Canada and Seoul (N=9), South Korea had a distinction in their quality of life (QoL). Physical environmental assessment was conducted using the Therapeutic Environment Screening Survey for Nursing Homes (TESS-NH) (Sloane et al., 2002). QoL was assessed three times over one year using Dementia Care Mapping tool (DCM) (University of Bradford, 2010). The results of the study demonstrated that the residents with dementia living in an institutional large-scale setting showed statistically more withdrawn behavior and spent more time to be negative mood or affect compared to the ones in a small-scale setting. This study also found that the number of potential positive behaviors of residents in a small-scale setting was three times higher than that of residents in an institutional large-scale setting. When looking at the distinction between two countries in the behavior category with a large average time difference, the residents with dementia in Korea had shorter meal/dessert times compared to those in Canada. The study supports that the small-scale homelike environment is intensely associated with a therapeutic environment for older adults with dementia.

Assessing quality and beneficial uses of Sargassum compost

Sargassum spp. (specifically Sargassum fluitans and S. natans), one of the dominant forms of marine macroalgae (seaweed) found on the beaches of Florida, is washing up on the shores throughout the Caribbean in record quantities. Currently, a common management option is to haul and dispose of beached Sargassum in local landfills, potentially wasting a valuable renewable resource. The objective of this study was to determine whether composting represents a feasible alternative to managing Sargassum inundations through measurements and comparisons to eleven guidelines. Specifically, we assessed the characteristics of the compost [physical–chemical parameters (temperature, moisture content, pH, and conductivity), nutrient ratios (C:N), elemental composition, bacteria levels, and ability to sustain plant growth] in both small- and large scale experiments. Results show that although nutrient concentration ratios were not within the standards outlined by the U.S. Composting Council (USCC), the Sargassum compost was able to sustain the growth of radishes (Raphanus sativus L., var. Champion). Trace metal concentrations in the compost product were within five regulatory guidelines evaluated, except for arsenic (As) (6.64–26.5 mg/kg), which exceeded one of the five (the Florida Soil Cleanup Target Level for residential use). Bacteria levels were consistent with regulatory guidelines for compost produced in large-scale outdoor experiments but not for the small-scale set conducted in enclosed tumblers. Overall results support that Sargassum compost can be beneficially used for fill and some farming applications.

Learning to empathize with users through design thinking in hybrid mode: Insights from two educational case studies

COVID-19 has imposed new educational challenges, including the need to prepare students to the new era of hybrid work. This research aimed to shed new light on the manner in which students learn and practice hybrid work in educational settings. To this end, two educational studies were performed, for examining students practicing design thinking (DT) in hybrid mode and the development of their empathy skills during this experience. The main case study was conducted in an academic hackathon-like course, where students addressed loneliness-related challenges during four full days using a configured web platform. In a follow-up case study, the adaptability of this approach to small-scale settings of a 90-minute DT workshop was explored. The students filled in questionnaires, which examined their empathy development in three dimensions: Skill, Orientation, and Being. The findings show the benefits of the web platform and the extent to which these academic educational programs fostered empathy and innovation and highlights the challenges, opportunities, and benefits of fostering new educational programs that allow hybrid participation, involving external stakeholders, and guide the students to handle real challenges. Such hybrid academic programs can prepare students for the hybrid, multidisciplinary work reality towards developing human-centered solutions.

DRL-Based Resource Allocation in Remote State Estimation

Remote state estimation where sensors send their measurements of distributed dynamic plants to a remote estimator over shared wireless resources is essential for mission-critical applications of Industry 4.0. Existing algorithms on dynamic radio resource allocation for remote estimation systems assumed oversimplified wireless communications models and can only work for small-scale settings. In this work, we consider remote estimation systems with practical wireless models over the orthogonal multiple-access and non-orthogonal multiple-access schemes. We derive necessary and sufficient conditions under which remote estimation systems can be stabilized. The conditions are described in terms of the transmission power budget, channel statistics, and plants’ parameters. For each multiple-access scheme, we formulate a novel dynamic resource allocation problem as a decision-making problem for achieving the minimum overall long-term average estimation mean-square error. Both the estimation quality and the channel quality states are taken into account for decision making. We systematically investigated the problems under different multiple-access schemes with large discrete, hybrid discrete-and-continuous, and continuous action spaces, respectively. We propose novel action-space compression methods and develop advanced deep reinforcement learning algorithms to solve the problems. Numerical results show that our algorithms solve the resource allocation problems effectively and provide much better scalability than the literature.

Use of the bias-corrected parametric bootstrap in sensitivity testing/analysis to construct confidence bounds with accurate levels of coverage

Sensitivity testing often involves sequential design strategies in small-sample settings that provide binary data which are then used to develop generalized linear models. Model parameters are usually estimated via maximum likelihood methods. Often, confidence bounds relating to model parameters and quantiles are based on the likelihood ratio. In this paper, it is demonstrated how the bias-corrected parametric bootstrap used in conjunction with approximate pivotal quantities can be used to provide an alternative means for constructing bounds when using a location-scale model. In small-sample settings, the coverage of bounds based on the likelihood ratio is often anticonservative due to bias in estimating the scale parameter. In contrast, bounds produced by the bias-corrected parametric bootstrap can provide accurate levels of coverage in such settings when both the sequential strategy and method for parameter estimation effectively adapt (are approximately equivariant) to the location and scale. A series of simulations illustrate this contrasting behavior in a small-sample setting when assuming a normal/probit model in conjunction with a popular sequential design strategy. In addition, it is shown how a high-fidelity assessment of performance can be attained with reduced computational effort by using the nonparametric bootstrap to resample pivotal quantities obtained from a small-scale set of parametric bootstrap simulations.

CNN-Based Transfer Learning for 3D Knuckle Recognition

A trustworthy and secure identity verification system is in great demand nowadays. The automatic recognition of the 3D middle finger knuckle is a new biometric identifier that could offer a precise, practical, and efficient alternative for personal identification. According to earlier studies, deep learning algorithms could be used for biometric identification. However, the accuracy of the current 3D middle finger knuckle recognition model is relatively low. Motivated by this fact, in this study, seven deep learning neural networks have been modified and trained to identify 3D middle finger knuckle patterns using transfer learning. Using the Hong Kong Polytechnic University’s 3D knuckle image dataset, an extensive experiment was performed. Two sessions of data from different camera lenses were used to assess the performance of the suggested deep learning model. The results show that the InceptionV3 method significantly enhanced the recognition of 3D middle finger knuckle patterns with 99.07% accuracy, followed by Xception, NasNetMobile, and DenseNet201 (97.35%, 92.92%, and 92.59%, respectively), which is superior to the current middle finger knuckle recognition model. This accurate, fast, and automatic middle finger knuckle identification will help to be implemented in real-time and small-scale settings like offices, schools, or personal devices like laptops and smartphones, where training is simple.

A key exchange system for secure data coordination in healthcare systems

To explore the coordination problem of the “gatekeeper” medical system, a referral game model between general hospitals and community hospitals within the system was established. Two non-competitive and competitive systems methods were modeled. The optimal referral threshold strategy and contract form under the service payment contract cooperation model. The conclusion shows that reducing the service cost of unit personnel in the system can effectively improve the system utility. Introducing competition can reduce the referral rate and promote the first consultation in community hospitals. The service payment contracts of non-competitive and competitive systems under different scales are feasible to achieve system coordination. The coordination efficiency of small-scale settings in competitive strategies is higher than that of large-scale environments. In the design of medical systems, competition cannot be avoided. Existing, setting up multiple small competing hospitals can bring higher social efficiency.

Distributed Optimization of District Heating Networks Using Optimality Condition Decomposition

The optimal operation of District Heating Networks (DHNs) is a challenging task. Current or future optimal dispatch energy management systems attempt to optimize objectives, such as monetary cost minimization, emission reduction, or social welfare maximization. Typically, this requires highly nonlinear models and has a substantial computational cost, especially for large DHNs. Consequently, it is difficult to solve the resulting nonlinear programming problem in real time. In particular, as typical applications allow for no more than several minutes of computation time. However, a distributed optimization approach may provide real time performance. Thereby, the solution of the central optimization problem is obtained by solving a set of small-scale, coupled optimization problems in parallel. At runtime, information is exchanged between the small-scale problems during the iterative solution procedure. A well-known approach of this class of distributed optimization algorithms is Optimality Condition Decomposition (OCD). Important advantages of this approach are the low amount of information exchange needed between the small-scale problems and that it does not require the tuning of parameters, which can be challenging. However, the DHNs model equation structure brings along many difficulties that hamper the application of the OCD approach. Simulation results demonstrate the applicability range of the presented method.

Abstract 547: Generation of synthetic cytokine receptor-induced cytotoxic innate lymphocytes (iCILs) from iPSCs as off-the-shelf cancer therapeutics

Abstract Off-the-shelf immune cell products derived from induced pluripotent stem cells (iPSC) have the potential to address many of the current limitations of engineered autologous and allogeneic blood-derived cell therapies. IPSC-based platforms offer a scalable and renewable source of allogeneic cells that can be engineered to augment cell phenotype and function. Current approaches to generating iPSC-derived immune effector cells have focused on recapitulating natural differentiation pathways to defined effector cell types (e.g., CD8 T cells or NK cells) utilizing feeder cells and exogenous cytokines. By engineering cells to express an artificial cytokine receptor, termed rapamycin activated cytokine receptor (RACR), we can control immune effector cell differentiation and expansion through synthetic signals independent of endogenous receptor expression, reducing the complexity and variability of cell differentiation while deriving cells with unique phenotypic and functional features not found in conventionally defined native immune cells. Here we demonstrate the ability to differentiate and expand synthetic cytotoxic innate lymphoid cells through engineering progenitor cells to express an artificial cytokine receptor, the RACR system, that is activated by a non-native small molecule ligand, rapamycin. This approach involves a period of feeder cell free hematopoietic and lymphoid progenitor specification followed by rapamycin-mediated activation of the artificial cytokine receptor (RACR), inducing an IL-2/IL-15-like signal to drive cytotoxic innate lymphocyte differentiation and feeder cell free expansion resulting in approximately 3000-fold expansion from the iPSC starting material. The protocol has demonstrated robustness in the small-scale setting with multiple research-grade iPSC lines and enables extensive effector cell expansion at low cost, enabled by the stability of the rapamycin ligand and elimination of complex raw materials such as exogenous cytokines. The resulting RACR-induced cytotoxic innate lymphocytes (RACR-iCILs) exhibit potent polyfunctional anti-tumor activity driven by the synergistic activity of innate immune receptors and the engineered expression of a chimeric antigen receptor (CAR), including cytolytic activity and the secretion of IFNγ and TNFα. The artificial cytokine receptor system can be used ex vivo to generate cells but also has the potential to enable rapamycin-controlled and selective expansion and survival of the engineered cells in vivo. Taken together, our data demonstrate the potential for de novo engineering of novel synthetic cytotoxic effector cells that show significant potential as “off the shelf” cancer therapeutics. We are currently moving these novel cells into humanized mouse models to further evaluate their unique properties and in vivo anti-tumor activity. Citation Format: Samantha O'Hara, Teisha Rowland, Ryan Koning, David Vereide, Michele Hoffmann, Ashley Yingst, Chris Nicolai, Mark Pankau, Kristen Mittelsteadt, Kathryn Michels, Seungjin Shin, Laurie Beitz, Byoung Ryu, Ryan Crisman, Andrew Scharenberg, Chris Garbe, Ryan Larson. Generation of synthetic cytokine receptor-induced cytotoxic innate lymphocytes (iCILs) from iPSCs as off-the-shelf cancer therapeutics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 547.

Using co-creation to facilitate PhD supervisory relationships

ABSTRACT The supervisory relationship is widely understood as central to the experience, success and wellbeing of PhD students. However, complex issues and struggles are frequently reported as associated with it. Although an extensive literature recommends useful, practical changes to improve supervisory relationships, current approaches generally focus on ameliorating difficulties within existing supervisory paradigms, rather than challenging or offering fresh perspectives on them. The co-creative approach has been successful in higher education, mostly at undergraduate level and in small-scale settings, providing opportunities for collaborative exploration of issues and development of solutions. Building on this, we explore the use of co-creation within the post-graduate research setting, bringing together stakeholders to unpack and address common issues experienced within the supervisory relationship. In order to assess the value of co-creation in tackling supervisory issues, we conducted a three-staged study involving PhD students and supervisors. Through qualitative and quantitative data (interviews and surveys), we identified specific issues linked to the supervisory experience. We subsequently hosted small- and large-scale co-creative workshops to encourage PhD students and supervisors to collaborate for addressing these identified issues. As a result, this paper argues that co-creation is an effective tool for enhancing the supervisory relationship and for co-solving problems associated with it. In addition, we present qualitative evidence supporting our novel use of the methodology and of the process of co-creation itself in addressing key identified issues within the supervisory relationship including isolation, a desire for community and improving communication skills.

HydroZen: The smart water tank system with advanced monitoring and automated refilling

The HydroZen smart water tank system is a modern solution for managing water supply in households and other smallscale settings. It uses sensors and Internet of Things (IoT) technology to monitor the water level and quality in real time and provides users with up-to-date information through a user-friendly mobile application. The system also incorporates automated refilling and flushing processes to ensure that the water remains at an acceptable level and quality. With its simple installation and user-friendly interface, the HydroZen smart water tank system offers an efficient and convenient way to manage water supply and promote responsible water usage.

Small-Scale Robust Digital Recognition of Meters Under Unstable and Complex Conditions

Digital recognition of meters aims to identify numbers in complex environments. Existing methods of digital recognition of meters are dependent on deep networks supported by high-quality large-scale data and features extraction, whereas low-quality small-scale digital datasets are not effective in recognition. Moreover, occlusion of digital images obstructs the extraction of fixed features. Multi-classifier under Feature Engineering (MC-FE) is proposed to solve the above problems. To be specific, MC-FE builds a feature library containing a variety of mainstream features. It directly selects the optimal combination of distinguishing features applicable to the current dataset, rather than using the fixed features. In addition, 10 regression machines integrated by support vector machines are adopted. The respective regression machine determines the probability of every number from 0 to 9. The positioning of the meter is the premise of accurate identification. A multi-layer kernel regression positioning (ML-KRP) is designed to increase the accuracy of meter identification. The results of the experiments on several digital recognition datasets reveal that MC- FE and ML-KRP outperform the state-of-the-arts in digital recognition under this small-scale setting.

Role of Physical Environment in Quality of Life among Residents in Dementia Care Facilities through a Longitudinal Observational Study : For Facilities in Canada and South Korea

This longitudinal observational study aims to examine whether residents with dementia in long-term care facilities with variability in physical environment qualities in Vancouver (N=11), Canada and Seoul (N=9), South Korea had a difference in their quality of life (QoL). Physical environmental assessment was conducted using the Therapeutic Environment Screening Survey for Nursing Homes (TESS-NH). QoL was assessed three times over one year using Dementia Care Mapping tool. The results of the study demonstrated that the residents with dementia living in an institutional large-scale setting showed statistically more withdrawn behavior and spent more time to be negative mood or affect compared to the ones in a small-scale setting. This study also found that the number of potential positive behaviors of residents in a small-scale setting was three times higher than that of residents in an institutional large-scale setting. When looking at the distinction between two countries in the behavior category with a large average time difference, the residents with dementia in Korea had shorter meal/dessert times compared to those in Canada. The study supports that the small-scale homelike environment is intensely associated with a therapeutic environment for older adults with dementia.

Predictive Representations in Hippocampal and Prefrontal Hierarchies.

As we navigate the world, we use learned representations of relational structures to explore and to reach goals. Studies of how relational knowledge enables inference and planning are typically conducted in controlled small-scale settings. It remains unclear, however, how people use stored knowledge in continuously unfolding navigation (e.g., walking long distances in a city). We hypothesized that multiscale predictive representations guide naturalistic navigation in humans, and these scales are organized along posterior-anterior prefrontal and hippocampal hierarchies. We conducted model-based representational similarity analyses of neuroimaging data collected while male and female participants navigated realistically long paths in virtual reality. We tested the pattern similarity of each point, along each path, to a weighted sum of its successor points within predictive horizons of different scales. We found that anterior PFC showed the largest predictive horizons, posterior hippocampus the smallest, with the anterior hippocampus and orbitofrontal regions in between. Our findings offer novel insights into how cognitive maps support hierarchical planning at multiple scales.SIGNIFICANCE STATEMENT Whenever we navigate the world, we represent our journey at multiple horizons: from our immediate surroundings to our distal goal. How are such cognitive maps at different horizons simultaneously represented in the brain? Here, we applied a reinforcement learning-based analysis to neuroimaging data acquired while participants virtually navigated their hometown. We investigated neural patterns in the hippocampus and PFC, key cognitive map regions. We uncovered predictive representations with multiscale horizons in prefrontal and hippocampal gradients, with the longest predictive horizons in anterior PFC and the shortest in posterior hippocampus. These findings provide empirical support for the computational hypothesis that multiscale neural representations guide goal-directed navigation. This advances our understanding of hierarchical planning in everyday navigation of realistic distances.

Large scale production of succinic acid by fermentation of sequential inorganic salt pretreated oil palm empty fruit bunch

The abundance of accumulated oil palm wastes has led to their usage as substrate in the production of organic acid such as succinic acid (SA). For the first time, oil palm empty fruit bunch (OPEFB) which was pretreated using inorganic salts was utilized for SA production in a large-scale fermenter. OPEFB was pretreated sequentially using 15% (w/v) Na3PO4.12H2O and 5% (w/v) ZnCl2 solution before undergoing fermentation process in a 16 L bioreactor in the presence of Actinobacillus succinogenes ATCC 55618. The optimum fermentation conditions were adopted from small-scale setting and implemented in the large-scale setting. Two different fermentation approaches were studied; simultaneous saccharification and fermentation (SSF) and modified simultaneous saccharification and fermentation (mSSF). This research revealed that mSSF produced higher concentration of total reducing sugars (TRS) yield and SA compared to the SSF method. SSF of inorganic salts pretreated OPEFB produced maximum SA concentration (38.85 gL-1) and yield (0.39 gg-1). Meanwhile highest TRS yield obtained from pre-hydrolysis was 0.79 gg-1 at 72 h and SA (50.5 gL-1) was acquired under the mSSF at 132 h. Therefore, mSSF is a novel process that has propitious potential for SA production in a large-scale setting which can be further executed in the industrial scale.