System Design Principles Research Articles

Achieving On-Site Trustworthy AI Implementation in the Construction Industry: A Framework Across the AI Lifecycle

In recent years, the application of artificial intelligence (AI) technology in the construction industry has rapidly emerged, particularly in areas such as site monitoring and project management. This technology has demonstrated its great potential in enhancing safety and productivity in construction. However, concerns regarding the technical maturity and reliability, safety, and privacy implications have led to a lack of trust in AI among stakeholders and end users in the construction industry, which slows the intelligent transformation of the industry, particularly for on-site AI implementation. This paper reviews frameworks for AI system design across various sectors and government regulations and requirements for achieving trustworthy and responsible AI. The principles for the AI system design are then determined. Furthermore, a lifecycle design framework specifically tailored for AI systems deployed in the construction industry is proposed. This framework addresses six key phases, including planning, data collection, algorithm development, deployment, maintenance, and archiving, and clarifies the design principles and development priorities needed for each phase to enhance AI system trustworthiness and acceptance. This framework provides design guidance for the implementation of AI in the construction industry, particularly for on-site applications, aiming to facilitate the intelligent transformation of the construction industry.

Physics-Based Modeling to Elucidate Design Principles for Redox-Mediated Electrochemical Systems

Electrochemical systems are increasingly being developed and deployed as sustainable technologies that support reductions in greenhouse gas emissions across the electric power, transportation, and industrial sectors. However, the need to overcome challenges related to cost, performance, and scalability continues to motivate exploration into alternative material sets and reactor designs that offer new pathways to widespread adoption. Redox-mediated processes, which decouple electrode reactions that activate a soluble mediator from “off-electrode” reactions where the activated mediator exchanges electrons with a solid material, are of growing interest.1,2 Recent literature has illustrated the potential of this approach for increasing the charge-storage capacity of redox flow batteries,2 performing impractical electrochemical processes,3 enhancing selectivity and efficiency of electrochemical separations and recycling,1,4,5 and unlocking spatial and temporal flexibility in electrochemical operations.6 Previously, we have developed and qualitatively validated a model capable of capturing key dynamics observed in redox-mediated flow batteries.7 Grounded in first principles and contextualized by experiments, this model has the potential to rapidly assess tradeoffs and opportunities for redox-mediated systems for applications in energy storage and conversion as well as materials recovery and recycling.Here, we describe how our model can be used to gain generalizable insights into the design-space for redox-mediated electrochemical processes. By incorporating models for pressure drop in porous media with simulated electrochemical performance, we assess the trade-offs associated with capacity utilization, power output, and parasitic pumping losses. Furthermore, we demonstrate how the model can support constrained optimization to guide the design of a system of known physical parameters. Ultimately, we seek to translate these findings into universal design principles for how electrochemical protocol and solid incorporation can be tuned to develop high-performing redox-mediated systems. Acknowledgements N.J.M gratefully acknowledges the NSF Graduate Research Fellowship Program under Grant Number 2141064. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. References Xue et al., Green Chem. 2020, 22 (19): 6288-6309.Yan and Wang, Mat. 2018, 30 (47).Vardner et al., ChemElectroChem 2022, 9 (24).Cotty et al., ACS Sustain Chem Eng. 2023, 11(9):3975-3986.Park et al., ACS Sustain Chem Eng. 2021, 9 (24): 8214-8221.Wang et al, Joule 2021, 5 (1): 149-165.Matteucci et al., ECS Meeting s 2023, MA2023-01 (3), 765.

Exploring the Role of Inert Conductive Area on Redox-Mediated Processes

To effectively utilize renewable energy resources, extensive exploration and investment has been directed towards advancing electrochemical technologies that enable reliable power delivery, vehicle electrification, and decarbonization of chemical manufacturing. Redox-mediated systems offer a pathway to decouple the electrochemical reactor from the conversion of an “off-electrode” material by introducing an activated mediator, potentially unlocking new processes opportunities.1 Recent literature highlights how this approach may increase capacity in redox flow batteries,2 facilitate difficult-to-perform electrochemical transformations,3 and allow spatial and temporal flexibility in electrochemical operations.4 However, numerous open questions still surround this concept, including how to predict the rates of the “off-electrode” reactions and how to tune design parameters to control and enhance these reactions. In earlier work, we showed mixed-potential theory could be used to understand performance trends observed in experimental studies of redox-mediated flow batteries.5 Intriguingly, the predicted reaction dynamics appear sensitive to the electronic conductivity of both the solid active materials and of the materials in direct contact with those charge-storage materials. Given the ubiquity of carbon-coated solid active materials found in redox-mediated flow batteries (e.g., lithium-ion intercalation materials2,6), interest in the use of redox-mediators to enhance metal-air batteries,7 and the overarching need to further elucidate design principles for redox-mediated electrochemical systems, understanding how conductive materials alter “off-electrode” processes is of interest.Here, through experiment and modeling, we explore how the incorporation of additional conductive but chemically inert surface area influences redox-mediated reactions between solid active materials and dissolved redox mediators. We pair electrochemical diagnostics with pre- and post-test materials characterization to assess the impact electron-conducting, inert surfaces in contact with the solid active material have on mediated reaction dynamics, extents, and distributions. In parallel, we develop a theoretical framework to aid in the interpretation of specific results and to generalize findings. Ultimately, these activities expand our knowledge of the role that solid-phase conductivity and conductive additives play in redox-mediated processes for energy storage and conversion. Acknowledgements N.J.M gratefully acknowledges the NSF Graduate Research Fellowship Program under Grant Number 2141064. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. References Qi & Koenig Jr., J. Vac. Sci. Technol. B, 2017, 35, 040801Jia et al., Sci Adv. 2015;1(10).Vardner et al., ChemElectroChem 2022, 9 (24).Anson & Stahl, Chemical Reviews, 2022, pp 3749–3786.Matteucci et al., ECS Meeting s 2023, MA2023-01 (3), 765.Jennings et al., Journal of Physical Chemistry C 2015, 119 (31), 17522–17528.Huang et al., Energy Environ Sci 2022, 16 (2), 438–445.

Sensing the shape of a surface by intracellular filaments.

The protein MreB, a homolog of eukaryotic actin, regulates the shape of bacteria like Escherichia coli by guiding new cell-wall insertion based on local curvature cues. However, the mechanism by which a nanometer-scale MreB filament "senses" the micron-scale curvature of the cell wall has remained a mystery. We introduce a biophysical model of the energetics of twisted and bent filaments bound to curved surfaces, which predicts that localization of filaments like MreB is sensitive to both mean and Gaussian curvature. The model captures experimentally measured curvature enrichment patterns and explains how MreB naturally localizes to saddle-shaped regions without energy-consuming processes. Beyond cell shape regulation, our work suggests design principles for synthetic systems that can sense and respond to surface shape.

Developing a Knowledge-Based Online Recommendation System to Guide STD Patients Through Their Journey

Sexually transmitted diseases (STDs) significantly impact public health, affecting one in five U.S. adults and imposing substantial economic burdens. Many at-risk individuals seek information and support online rather than through regular testing, facing challenges due to fragmented information and a lack of personalized recommendations. This study develops an online health recommendation system (OHRS) tailored for STD patients, integrating informational and emotional support based on their disease journey. Using a BERT-based named entity recognition (NER) algorithm, the system identifies patient emotions and stages from online posts, providing relevant support resources. Data collection for designing a recommendation engine included analyzing online posts and consulting healthcare experts. The system's effectiveness was validated through user-based simulation studies. Key contributions include developing text-mining algorithms, creating a knowledge-based recommendation system, and proposing design principles for AI-driven support systems for stigmatized conditions.

The need for general adaptive capacity-Discussing resilience with complex adaptive systems theory.

The concept of resilience intrinsically links with both complexity and adaptive capacity. Scholars from different fields agree on this. Still, the detailed relations between resilience, complexity, and adaptive capacity need a more thorough theoretical analysis. This article analyses resilience with the help of assumptions from complex adaptive systems (CAS) theory to answer two questions in more detail: What is the relation between resilience and complexity? How can adaptive capacity contribute to resilience? By applying basic ideas from CAS theory to the resilience discourse, the article deduces that complexity of a system is a necessary condition for resilience because complex systems consist of agents that possess adaptive capacity, whereas simple systems consist of mere elements that cannot adapt to unexpected disruptions. The relation between complexity and resilience is multidimensional. Growing complexity leads to a growing need for resilience because the chances for severe, unexpected disruptions increase. The analysis of adaptive capacities revealed that systems and the agents they consist of can possess of specialized and general adaptive capacity. General adaptive capacity is the core feature of resilience because it enables systems to cope with unexpected disruptions. System design principles such as diversity within functional groups and redundancy help to increase general adaptive capacity. The same is true on the community level for social capital and on the individual level for disaster preparedness measures because they increase coping capacities independent of specific hazards.

Persuasive software features as antecedents to perceived compliance within physical activity behavior change support systems

ABSTRACT Although behavior change support systems (BCSS) have the potential to improve compliance using persuasive software features, there is a dearth of empirical evidence on how these features impact perceived compliance. This study aimed to investigate how and to what extent the persuasiv systems design (PSD) principles in physical activity BCSS influence perceived compliance. Quantitative data from 403 participants (165 users and 238 prospective users of BCSS designed for PA) was analyzed using partial least squares structural equation modeling. Results revealed differences in how the PSD principles impact perceived compliance between users and prospective users. Perceptions of dialogue support, credibility support, and social support were more relevant to prospective users than to users. Thus, designers should pay more attention to these principles to attract and engage prospective users of BCSS. Conversely, primary task support was more relevant to users than to prospective users. This indicates that once prospective users transition to active users, the focus should shift towards providing persuasive features that facilitate the performance of the primary task. These findings underscore the importance of developing intelligent BCSS that can adapt to the evolving needs and preferences of users at different stages of their interactions with a system.

A Study on the Impact of Quality and HCI Factors of a Deep Learning-Based Makeup Recommendation System on Acceptance Intention: Focusing on the Mediating Effects of Users’ Perceived Trust and Immersion

This study analyzed the impact of system quality and HCI factors on user acceptance intention, mediated by perceived trust and immersion, in the context of a deep learning-based makeup recommendation system. The results indicate that system quality and HCI factors significantly influence user acceptance intention through perceived trust and immersion. This underscores the importance of system quality and HCI factors for the successful application of makeup recommendation systems. In particular, factors such as diversity, accuracy, anthropomorphism, and realism enhance user trust and immersion, leading to higher acceptance intention. However, it was found that certain factors, such as personalization, do not significantly influence acceptance intention through immersion. This suggests that the impact of specific factors should be considered in the design of recommendation systems. The study is limited by its focus on a specific age group and gender in the survey, which may restrict the generalization of the results. Additionally, there is a lack of research on users with different cultural backgrounds, necessitating an understanding of the global market applicability. Future research should extend to users with diverse demographic characteristics to derive universal design principles for makeup recommendation systems. Furthermore, the impact of factors such as personalization on user experience should be explored in depth to enhance user-centered, customized beauty services. This research provides a theoretical and practical foundation for the effective design and implementation of makeup recommendation systems and is expected to contribute to the digital transformation of the beauty industry.

Epistemic Injustice in Generative AI

This paper investigates how generative AI can potentially undermine the integrity of collective knowledge and the processes we rely on to acquire, assess, and trust information, posing a significant threat to our knowledge ecosystem and democratic discourse. Grounded in social and political philosophy, we introduce the concept of generative algorithmic epistemic injustice. We identify four key dimensions of this phenomenon: amplified and manipulative testimonial injustice, along with hermeneutical ignorance and access injustice. We illustrate each dimension with real-world examples that reveal how generative AI can produce or amplify misinformation, perpetuate representational harm, and create epistemic inequities, particularly in multilingual contexts. By highlighting these injustices, we aim to inform the development of epistemically just generative AI systems, proposing strategies for resistance, system design principles, and two approaches that leverage generative AI to foster a more equitable information ecosystem, thereby safeguarding democratic values and the integrity of knowledge production.

Digital behaviour change intervention for weight loss maintenance in adults with obesity: a feasibility pilot study of eCHANGE

ABSTRACT Successful weight maintenance after weight loss is challenging. Digital interventions may facilitate behaviour change to prevent weight regain, however little is known about the acceptance and use of digital interventions in support of weight maintenance. This mixed methods study aims to evaluate user experiences, system use, and preliminary efficacy of eCHANGE, an application based self-management intervention for weight loss maintenance. A 3-month multi-site single group feasibility pilot study was conducted among adults (N = 60) with BMI ≥30 kg/m2, aiming to maintain weight after weight loss of ≥8%. User experiences and system use were examined through validated questionnaires, system use log data, and individual interviews (n = 15). Preliminary efficacy testing included body weight and patient reported outcome measures. Participants rated eCHANGE usability and usefulness as good in support of weight maintenance, with variation in usage. Analysis indicated that higher behavioural engagement scores were statistically significantly associated with frequency of technology usage. Weight loss was maintained by 83% of the participants at 3-months (i.e. defined as weight change of <3% of baseline body weight (kg)). Digital interventions, combining persuasive system design principles and behaviour change techniques, supporting self-regulation and maintenance of health behaviours, have the potential to facilitate weight maintenance after weight loss. Trial Registration: ClinicalTrials.gov NCT04537988 https://clinicaltrials.gov/ct2/show/NCT04537988

From Pursuit of the Universal AGI Architecture to Systematic Approach to Heterogeneous AGI (SAGI): Addressing Alignment, Energy &amp; AGI Grand Challenges

Artificial intelligence (AI) faces a trifecta of grand challenges: the Energy Wall, the Alignment Problem and the Leap from Narrow AI to AGI. Contemporary AI solutions consume unsustainable amounts of energy during model training and daily operations. Making things worse, the amount of computation required to train each new AI model has been doubling every 2 months since 2020, directly translating to unprecedented increases in energy consumption. The leap from AI to AGI requires multiple functional subsystems operating in a balanced manner, which requires a system architecture. However, the current approach to AI lacks system design; even though system characteristics play a key role in the human brain; from the way it processes information to how it makes decisions. In this paper, we posit that system design is the missing piece in overcoming current AI the grand challenges. We present a Systematic Approach to AGI (SAGI) that utilizes system design principles to overcome the energy wall and the alignment challenges. This paper asserts that artificial intelligence can be realized through a multiplicity of design-specific pathways, rather than a singular, overarching AGI architecture. AGI systems may exhibit diverse architectural configurations and capabilities, contingent upon their intended use cases. We argue that AI alignment, the most difficult among the grand challenges, is not attainable without a way to reflect the complexity of the human moral system and its subsystems in the AGI architectures. We claim that AGI approaches such as symbolicism, connectionism and others are not fundamental to AGI but emergent from the system design processes. Hence, we focus on employing system design principles as a guiding framework, rather than solely concentrating on a universal AGI architecture.

Designing social media to foster user engagement in challenging misinformation: a cross-cultural comparison between the UK and Arab countries

Challenging others who post misinformation is a type of social correction that complements algorithm-based approaches. However, participation rates in such social acts remain limited. In this paper, we study design techniques that leverage principles of persuasive system design and communication theories to foster such prosocial behaviour across two distinct cultural contexts: the British and the Arab. A total of 462 participants completed an online survey (250 UK, 212 Arabs). The study compared the two cultural contexts regarding willingness to challenge misinformation and the persuasiveness of seven design techniques to increase that willingness, namely predefined question stickers, thinking face reaction, sentence openers, fact checker badge, social norm messages, tone detector, and private commenting. Moreover, it explores the impact of individuals’ characteristics on their perception of the techniques as being more or less persuasive than a standard comment box. The study found that the willingness to challenge misinformation was significantly higher in the Arab context than in the UK context. Moreover, except for the private commenting, all techniques were more impactful in the Arab context than in the UK context. Some techniques, such as predefined question stickers, were more effective in both cultures compared to the standard comment box, while others, like the fact checker badge, were more effective only in the Arab context. However, in the UK, sentence openers had a lower impact. Furthermore, personality traits, age, and perspective-taking showed the potential but also the varying impacts on the persuasiveness of the techniques on users’ correction of misinformation across both cultural contexts while pointing to the need for considering both personal and cultural factors in designing social-correction-based solutions.

Taming Consensus in the Wild (with the Shared Log Abstraction)

The shared log is an abstraction for building layered consensus systems that are simple to develop, deploy, evolve, and operate. Shared logs emerged from systems research and have seen significant traction in industry over the past decade. In this paper, we describe some design principles for consensus-based systems, based on our experience building and operating real-world shared log databases in the wild.

Electrochemically-Driven Reversible Catch and Release of Rare Earth Elements

The sustainable recovery of critical materials, such as rare earth elements (REEs), from dilute and non-traditional sources is pivotal for securing a reliable domestic supply. Using environmentally friendly methods is necessary to maximize the economic and sustainability benefits. One of the most promising approaches is the electrochemically controlled recovery of REEs, which becomes even more compelling when integrated with renewable energy sources. We developed a groundbreaking electroactive material with an exceptional affinity for REEs in their trivalent state. This material demonstrates a remarkable ability for highly reversible binding and release of REEs, controlled by the redox state of the material. We will present the REE recovery system's design principles and performance characteristics, mainly focusing on its efficacy in extracting REEs from dilute sources and structure-property relationships.

Competitive Valerate Binding Enables RuO2-Mediated Butene Electrosynthesis in Water.

The (non)-Kolbe oxidation of valeric acid, sourced from a hydrolysis product of cellulose, provides a sustainable synthetic route to access value-added products, such as butene. An essential mechanistic step preceding product formation involves the oxidative and decarboxylative cleavage of a C-C bond. Yet, the role of the electrode surface in mediating this oxidative step remains an open question: the electron transfer can occur either via an inner-sphere or outer-sphere mechanism. Here, we report the electrochemical, in situ spectroscopic, computational, and reactivity studies of RuO2-mediated oxidative decarboxylation of valeric acid to butene in aqueous electrolytes. We find that carboxylates bind to RuO2 anode surfaces at potential values where decarboxylation products are observed. Our results are consistent with a reaction scheme where the competitive and catalytic oxygen evolution reaction (OER) is impeded by these bound carboxylate species while these species are inert toward butene formation. Our results implicate an outer-sphere electron transfer mechanism for decarboxylation where the surface chemistry of the RuO2 electrode serves to enable higher non-Kolbe reaction selectivity by suppressing the parasitic OER. Our findings delineate interfacial design principles for selective electrochemical systems that utilize water as the ultimate oxidant for sustainable decarboxylation.

Turning Data Center Waste Heat into Energy: A Guide to Organic Rankine Cycle System Design and Performance Evaluation

This study delves into the adoption of the organic Rankine cycle (ORC) for recovering waste heat from data centers (DCs). Through a literature review, it examines energy reuse with a focus on electric power generation, the selection of working fluids, and system design principles. The objective is to develop a thorough framework for system design and analysis, beginning with a quantity and quality investigation of waste heat available. Air cooling systems, chosen often for their simplicity, account for about 70% of used cooling methods. Water cooling demonstrates greater effectiveness, albeit less commonly adopted. This study pays close attention to the selection of potential working fluids, meticulously considering the limitations presented by the available sources of heat and cold for vaporization and condensation, respectively. It reviews an ORC-based system setup, incorporating fluid streams for internal processes. The research includes a conceptual case study where the system is designed and simulations are conducted in the DWSIM environment. The simulation model considers hot air or hot liquid water returning from the data center cooling system for ORC working fluid evaporation. Ambient water serves for condensing, with pentane and isopentane identified as suitable organic fluids. Pentane assures ORC net electric efficiencies ranging between 3.1 and 7.1% when operating pressure ratios increase from 2.8 to 6.4. Isopentane systems, meanwhile, achieve efficiencies of 3.6–7.0% across pressure ratios of 2.7–6.0. Furthermore, the investigation provides key performance indicators for a reference data center in terms of power usage effectiveness (PUE), energy reuse factor (ERF), energy reuse effectiveness (ERE), and greenhouse gas (GHG) savings. This study concludes with guidelines for system analysis, including exergy considerations, and details the sizing process for evaporators and condensers.

Scaling laws for optimal power-law fluid flow within converging–diverging dendritic networks of tubes and rectangular channels

Flows in dendritic–fractal networks have garnered extensive research attention, but most studies assume a constant tube or channel cross section. In many applications, the cross section of the tube or channel changes as the flow progresses through it, such as the blood flow through the arterial system, which is a prime example of a deformable or non-uniform tree-like network. Heating, ventilation, and air conditioning ductwork also exemplify a tree-like network with varying cross sections. This research investigates power-law fluid flows in the converging–diverging tubes and rectangular channels, prevalent in engineered microfluidic devices, many industrial processes, and heat transfer applications. Power-law fluid flows through linear, parabolic, hyperbolic, hyperbolic cosine, and sinusoidal converging–diverging dendritic networks of tubes and rectangular channels are studied. The flow is assumed to be steady, incompressible, two-dimensional planar, and axisymmetric laminar flow without considering network losses. A theoretical model has been derived to evaluate the flow conductance under network volume and surface-area constraints. The flow conductance is highly sensitive to network geometry. The effective conductance of all networks increases with increasing daughter-to-parent radius ratio before eventually declining. The maximum conductance occurs when a specific radius or channel-height daughter–parent ratio β* is achieved. This value depends on the constraint and vessel geometry, such as tubes or rectangular channels. The optimal flow conditions for maximum conductance in a constrained tube volume network, βmax*=βmin*=N−1/3, while for a constrained tube's surface-area network, βmax*=βmin*=N−(n+1)/(3n+2). This scaling applies to all converging–diverging tube network profiles. Here, βmax*, βmin* are the radius ratios of the daughter–parent pair at the maximum divergent or minimum convergent part of the vessel. N represents the number of branches splitting at each junction, and n is the power-law index of the fluid. Furthermore, the optimal flow scaling for the height ratio in the rectangular channel, βmax*=βmin*=N−1/2α−1/2 for constrained channel volume and βmax*=βmin*=N−1/2α−n/(2n+2) for constrained surface area for all converging–diverging channel networks, respectively, where α is the channel-width ratio between parent and daughter branches. Additionally, at optimal conditions in both the channels and tube network, pressure drops are equally partitioned across each branching level. The results in this work are validated with experiments and existing theories for limiting conditions. This research expands existing design principles for efficient flow systems, previously in the literature developed for uniform vessels, to encompass non-uniform converging–diverging vessels. Additionally, it provides a valuable framework for studying non-Newtonian flows within complex, non-uniform tree-like networks.

Reconsidering Leadership Development: From Programs to Developmental Systems.

We argue for reconsidering leadership development based on open systems theory and systems design principles. A primary advantage of open systems thinking is that it encourages holistic approaches to development and avoids episodic program-based training and piecemeal thinking. The latter approaches are both limited and limiting yet tend to be the preferred approach to leadership development in organizations. Open systems approaches to development offer numerous advantages both conceptually and pragmatically, especially through the incorporation of ongoing feedback cycles. Core practices that define a leadership development system are presented and implications are discussed.

46 Principles of industrial ventilation system design and operation

46 Principles of industrial ventilation system design and operation

Ether-Based High-Voltage Lithium Metal Batteries: The Road to Commercialization.

Ether-based high-voltage lithium metal batteries (HV-LMBs) are drawing growing interest due to their high compatibility with the Li metal anode. However, the commercialization of ether-based HV-LMBs still faces many challenges, including short cycle life, limited safety, and complex failure mechanisms. In this Review, we discuss recent progress achieved in ether-based electrolytes for HV-LMBs and propose a systematic design principle for the electrolyte based on three important parameters: electrochemical performance, safety, and industrial scalability. Finally, we summarize the challenges for the commercial application of ether-based HV-LMBs and suggest a roadmap for future development.