Integrating lean maintenance and smart monitoring to enhance energy efficiency in hybrid solar-mechanical systems

Corresponding author: Modupe Arowolo Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. Integrating lean maintenance and smart monitoring to enhance energy efficiency in hybrid solar-mechanical systems Modupe Arowolo *, Oluremi Funmilayo Hamid and Marvin Baptiste Department of Engineering Technology, Western Illinois University, USA. World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 Publication history: Received on 12May 2025; revised on 18 June 2025; accepted on 20 June 2025 Article DOI: https://doi.org/10.30574/wjarr.2025.26.3.2387 Abstract This article examines the integration of lean maintenance methodologies with smart monitoring technologies to optimize energy efficiency in hybrid solar-mechanical systems across the United States. Through empirical analysis and case studies, we demonstrate how principles such as 5S, Kaizen, and PDCA can be effectively combined with IoT sensors, machine learning algorithms, and predictive analytics to reduce waste, minimize downtime, and maximize energy yield. Our findings indicate that integrated approaches can achieve energy efficiency improvements of 15-27% compared to traditional maintenance regimes, with corresponding reductions in operational costs of 18-32%. The research highlights the synergistic relationship between process optimization and technological innovation, providing a framework for implementation that is adaptable across various scales and configurations of hybrid energy systems. Keywords: Lean maintenance; Hybrid solar-mechanical; IoT sensors; Machine learning; Synergistic 1. Introduction The U.S. renewable energy sector has experienced unprecedented growth, with solar capacity increasing by 25% annually over the past five years (DOE, 2024). Hybrid solar-mechanical systems—which combine photovoltaic generation with mechanical components such as HVAC systems, compressors, and pumps—represent a growing segment of this market. These systems offer compelling advantages, including enhanced reliability, improved load management, and increased overall efficiency. However, they also present unique maintenance challenges due to their heterogeneous components and operational complexity. Traditional maintenance approaches often fail to address the specific requirements of these hybrid systems, resulting in suboptimal performance, premature component failure, and energy losses. Simultaneously, the emergence of smart monitoring technologies presents new opportunities for real-time system oversight and predictive maintenance. This article explores how lean maintenance philosophies can be integrated with these technological innovations to create more efficient, reliable, and sustainable hybrid energy systems. 1.1. Research Objectives •To develop a comprehensive framework for integrating lean maintenance principles with smart monitoring technologies in hybrid solar-mechanical systems •To quantify the energy efficiency improvements resulting from this integrated approach •To identify key implementation challenges and success factors in U.S. installations •To establish best practices for scaling these solutions across different system configurations and operational environments. World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1982 2. Literature Review 2.1. Lean Maintenance in Energy Systems Lean maintenance, derived from Toyota's production system, focuses on eliminating waste and maximizing value (Womack and Jones, 2003). In energy systems, this philosophy has been applied to reduce downtime, optimize resource utilization, and enhance overall operational efficiency. Key lean methodologies relevant to hybrid solar-mechanical systems include: • 5S (Sort, Set in order, Shine, Standardize, Sustain): Creates organized, efficient workspaces that reduce time spent locating tools and components during maintenance activities • Kaizen: Promotes continuous improvement through incremental changes to maintenance processes • Total Productive Maintenance (TPM): Emphasizes proactive and preventive maintenance to maximize equipment effectiveness • Plan-Do-Check-Act (PDCA): Provides a structured approach to problem-solving and process improvement Recent studies have demonstrated the effectiveness of lean maintenance in renewable energy applications. For example, Patel et al. (2023) reported a 22% reduction in maintenance costs and a 17% improvement in overall equipment effectiveness (OEE) after implementing lean principles at a 50 MW solar facility in Arizona. 2.2. Smart Monitoring Technologies The rapid advancement of Internet of Things (IoT) technologies, artificial intelligence, and data analytics has transformed maintenance capabilities in energy systems. Smart monitoring encompasses: • Networked sensors that continuously monitor system parameters • Data transmission infrastructure • Analytics platforms that process and interpret data • Visualization tools that present actionable insights • Automated response mechanisms These technologies enable: • Real-time performance monitoring across distributed system components • Fault detection and diagnosis with increasing accuracy and decreasing false alarms • Predictive maintenance that anticipates failures before they occur • Performance optimization through continuous data analysis and feedback According to a comprehensive survey by the National Renewable Energy Laboratory (NREL, 2024), smart monitoring technologies have reduced unplanned downtime by 35-45% in solar installations across the U.S. Southwest. 2.3. Hybrid Solar-Mechanical Systems Hybrid solar-mechanical systems integrate photovoltaic generation with mechanical components to create more versatile and efficient energy solutions. Common configurations include: • Solar PV combined with HVAC systems for commercial buildings • Solar-powered compressors for industrial applications • Integrated solar and mechanical pumping systems for agriculture • Solar PV with mechanical energy storage (compressed air, flywheels) These systems present unique maintenance challenges due to their: • Diverse component types with different maintenance requirements • Complex interactions between electrical and mechanical subsystems • Varied degradation mechanisms affected by environmental conditions • Multiple potential failure modes. World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1983 3. Integration Framework 3.1. Conceptual Model The integration of lean maintenance and smart monitoring for hybrid solar-mechanical systems requires a comprehensive framework that addresses both philosophical and technological elements. Figure 1 presents the conceptual model developed through this research. Figure 1 Conceptual model for integrating lean maintenance and smart monitoring in hybrid solar-mechanical systems The model illustrates how lean principles provide the philosophical foundation, while smart technologies deliver the data and insights necessary for implementation. The integration occurs through: • Synchronized data flows that connect physical systems to decision processes • Standardized workflows that incorporate both automated and manual elements • Feedback mechanisms that enable continuous improvement • Decision support systems that translate data into actionable maintenance directives 3.2. Key Components The successful integration of lean maintenance and smart monitoring in hybrid solar-mechanical systems relies on several key components: 3.2.1. Sensor Infrastructure A comprehensive sensor network forms the foundation of the integrated approach. Essential sensor types include: • Irradiance sensors to measure solar resource availability • Temperature sensors at multiple system points • Vibration sensors on rotating mechanical components • Electrical performance sensors (voltage, current, power quality) World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1984 • Fluid flow and pressure sensors for hydraulic components • Acoustic sensors for early failure detection Table 1 summarizes the typical sensor requirements for different system sizes. Table 1 Sensor Requirements by System Size System Size PV Capacity Mechanical Components Minimum Sensor Count Data Points/Day Storage Requirements Small <100 kW 1-2 25-50 1.8-3.6 million 15-30 GB/year Medium 100-500 kW 3-5 50-150 3.6-10.8 million 30-90 GB/year Large >500 kW 6+ 150-500+ 10.8-36 million+ 90-300+ GB/year 3.2.2. Data Architecture An effective data architecture must balance: • Edge computing for real-time processing and immediate response • Cloud integration for advanced analytics and pattern recognition • Secure data transmission and storage • Scalable infrastructure that can grow with system expansion • Interoperability between diverse hardware and software platforms 3.2.3. Analytics Capabilities Modern analytics capabilities essential for the integrated approach include: • Descriptive analytics that characterize system performance • Diagnostic analytics that identify root causes of deviations • Predictive analytics that forecast potential failures • Prescriptive analytics that recommend optimal maintenance actions 3.2.4. Visualization and Reporting Effective visualization tools enable: • Real-time dashboards for operators • Trend analysis for maintenance planners • Exception reports for management • Mobile interfaces for field technicians 3.2.5. Integration with Lean Workflows The technological elements must be seamlessly integrated with lean maintenance workflows: • 5S principles applied to both physical maintenance areas and digital workspaces • Standardized procedures that incorporate data insights • Visual management that combines digital displays with traditional visual controls • Continuous improvement mechanisms that leverage both human expertise and AI-driven insights. 4. Implementation Methodology 4.1. Phased Approach Our research team developed and validated a comprehensive phased implementation methodology that balances immediate operational gains with long-term transformational objectives. Rather than attempting a wholesale system overhaul—which often leads to resistance and implementation failure—we structured the integration of lean World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1985 maintenance and smart monitoring technologies as a progressive journey. This approach allowed organizations to build capabilities incrementally, demonstrate value early, and adjust strategies based on emerging insights. "The most critical success factor we identified was proper sequencing," notes Dr. James Wilson, lead implementation specialist for the Phoenix case study. "Organizations that attempted to deploy advanced analytics before establishing reliable sensor infrastructure and standardized workflows invariably struggled with data integrity issues and low adoption rates." The methodology consists of four distinct phases, each with specific objectives, activities, and deliverables: 4.1.1. Phase 1: Foundation Building (1-3 months) The initial phase focuses on establishing the technical and organizational foundation required for successful integration. We began by conducting comprehensive baseline performance assessments across all hybrid solar-mechanical systems to establish clear metrics against which future improvements could be measured. This involved detailed energy efficiency testing, maintenance cost analysis, and reliability assessments using internationally recognized standards such as ISO 50001 and IEEE 493. At the Phoenix commercial office site, we deployed 127 sensors strategically positioned throughout both solar and mechanical components, creating a high-resolution monitoring network. Sensor selection and placement required careful consideration of environmental conditions, particularly for outdoor components exposed to Arizona's extreme temperature variations. We utilized NEMA-rated enclosures with passive cooling systems to ensure sensor reliability in ambient temperatures exceeding 110°F during summer months. Data collection architecture implementation followed, with edge computing devices installed at five strategic locations to perform initial data processing before transmission to the central monitoring platform. This approach reduced bandwidth requirements by approximately 73% while enabling sub-second response times for critical parameters. The team developed custom visualization dashboards with both technical and management-oriented views to ensure data accessibility across different organizational roles. Concurrent with the technical deployment, we implemented initial 5S activities in maintenance areas. This involved: • Sort: Removing unnecessary tools, parts, and documentation from maintenance facilities • Set in order: Establishing dedicated storage locations for equipment, with priority given to frequently used items • Shine: Developing regular cleaning protocols for both maintenance areas and system components • Standardize: Creating visual reference guides for proper workspace organization • Sustain: Implementing daily audit protocols to maintain organizational standards The 5S implementation required approximately 140 person-hours across three maintenance teams but reduced tool retrieval time by 67% and eliminated an estimated 15 hours of monthly search time previously spent locating parts and documentation. 4.1.2. Phase 2: Process Integration (3-6 months) With the foundation in place, Phase 2 focused on integrating the newly available data streams with operational maintenance processes. This phase proved particularly challenging as it required significant procedural changes for maintenance personnel who had established routines developed over many years. We began by mapping existing maintenance workflows using value stream mapping techniques, identifying 12 critical maintenance processes and 37 subprocesses across the hybrid systems. Through collaborative workshops with maintenance teams, we identified numerous opportunities for waste reduction, including excessive travel time between components, duplicative inspection activities, and preventive maintenance tasks performed at unnecessarily high frequencies. The development of standardized maintenance procedures incorporated insights from both historical maintenance records and newly available sensor data. For example, at the Phoenix site, analysis of vibration patterns from air handler units revealed that preventive maintenance intervals could be safely extended from 30 to 45 days, reducing annual maintenance hours by approximately 120 while maintaining system reliability. Conversely, data from the solar inverters World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1986 indicated that more frequent inspection of DC connections was warranted due to thermal cycling effects in the desert environment. Integration of sensor data with maintenance workflows required careful attention to information delivery mechanisms. We implemented a tiered notification system with clear response protocols: • Level 1 (Information): System parameters outside nominal range but within acceptable limits; documented for trend analysis • Level 2 (Advisory): Parameters approaching action thresholds; scheduled for evaluation during next maintenance cycle • Level 3 (Warning): Parameters requiring attention within defined timeframe (24-72 hours) • Level 4 (Critical): Parameters requiring immediate intervention Training of maintenance personnel represented a significant investment during this phase, with each technician receiving 32 hours of formal instruction and an additional 40 hours of on-the-job training with senior mentors. We developed a comprehensive curriculum covering lean maintenance principles, data interpretation, tablet-based maintenance management, and newly standardized procedures. Training effectiveness was evaluated through practical assessments rather than traditional testing, with technicians demonstrating proficiency in real-world scenarios. The implementation of visual management systems provided the "central nervous system" connecting sensor data with human action. Large-format digital displays were installed in maintenance areas, providing real-time system status information with intuitive visual cues. Color coding, standardized iconography, and progressive disclosure of information ensured that technicians could quickly assess system status and prioritize activities appropriately. 4.1.3. Phase 3: Advanced Analytics (6-12 months) The third phase leveraged the now-reliable data streams and standardized processes to implement increasingly sophisticated analytical capabilities. This phase represented the transition from reactive and preventive maintenance to truly predictive and eventually prescriptive approaches. Deployment of predictive maintenance algorithms began with relatively straightforward applications focused on wellunderstood failure modes. For example, we implemented bearing failure prediction models using vibration signature analysis for rotating equipment, achieving 87% accuracy in detecting incipient failures approximately 45 days before functional impact. This provided ample time for planned interventions during regular maintenance windows. As confidence in the basic algorithms grew, we progressively implemented more complex predictive models targeting system-level performance rather than individual component failures. These included: • Solar yield deviation detection accounting for environmental variables (irradiance, temperature, soiling) • HVAC efficiency degradation models incorporating building occupancy and weather forecasts • Integrated energy flow optimization across the solar-mechanical interface The development of prescriptive maintenance recommendations represented a particularly valuable advancement. Rather than simply alerting maintenance personnel to developing issues, the system began providing specific recommended actions prioritized by impact, resource requirements, and scheduling constraints. At the Denver site, this capability reduced maintenance planning time by approximately 7 hours per week while improving intervention effectiveness. Integration with enterprise asset management systems required significant effort to ensure bidirectional data flow between the monitoring platform and existing maintenance management software. We developed custom APIs to facilitate this integration, enabling automated work order generation, parts inventory management, and maintenance history tracking. This integration eliminated approximately 15 hours of weekly administrative burden previously required for manual data entry and report generation. The establishment of continuous improvement mechanisms created the framework for ongoing refinement beyond the initial implementation. We implemented monthly review sessions bringing together maintenance, operations, and management personnel to analyze system performance, identify improvement opportunities, and develop action plans. These sessions utilized standardized A3 problem-solving methodology to ensure structured analysis and documentation. World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1987 4.1.4. Phase 4: Optimization and Scaling (Ongoing) The final phase focuses on ongoing refinement and expansion of the integrated approach. Unlike the preceding phases, Phase 4 has no definitive end point, instead representing a commitment to continuous evolution and improvement. Refinement of algorithms based on accumulated operational data has proven particularly valuable. As the systems accumulated 18+ months of operational data across seasonal variations, we were able to retrain the predictive models with substantially larger datasets, increasing prediction accuracy by approximately 12% compared to the initial models. The incorporation of transfer learning techniques allowed insights gained at one site to improve model performance across all installations. Expansion to additional systems and sites follows a standardized methodology developed through earlier implementations. At the Charlotte campus, we successfully expanded from the initial implementation on the science building to seven additional structures using a streamlined approach that reduced per-building implementation time by approximately 40% compared to the original deployment. This expansion methodology included: • Standard sensor packages tailored to three building categories (administrative, instructional, research) • Pre-configured analytics modules with site-specific parameter adjustment • Accelerated training program for maintenance personnel leveraging existing expertise • Templated integration configurations for common building management systems • The development of advanced optimization strategies represents the frontier of current implementation efforts. Cross-system optimization algorithms now enable coordinated operation of solar production, energy storage, and mechanical systems based on real-time pricing signals, weather forecasts, and predicted occupancy patterns. At the Phoenix site, these algorithms have reduced peak demand charges by approximately 23% while maintaining or improving occupant comfort metrics. • Knowledge sharing across the organization has proven essential for sustaining and expanding benefits. We established a digital repository of implementation learnings, including case studies, problem-solving examples, and technical guidance. Monthly virtual collaboration sessions bring together maintenance teams from different sites to share experiences and best practices. This knowledge ecosystem has accelerated innovation while reducing implementation barriers for subsequent deployments. • "The phased approach we developed provides a structured yet flexible pathway for organizations to transform their maintenance practices," explains Maria Rodriguez, systems integration lead. "By balancing technical implementation with organizational change management, we've achieved sustainable improvements that continue to evolve and expand well beyond the initial implementation timeline." • The methodology detailed above has been successfully implemented across diverse organizational contexts, facility types, and geographic locations. While specific implementation timelines and emphasis areas may vary based on organizational readiness and priorities, the fundamental phased structure has proven robust and adaptable. 4.2. Case Study: Commercial Office Building in Phoenix, Arizona To validate the integrated approach, we implemented the framework at a commercial office complex in Phoenix, Arizona, featuring: • 350 kW rooftop solar PV system • Integrated solar-assisted HVAC system • Multiple air handling units and compressors • Building automation system The implementation followed the phased approach described above, with key activities including: • Installation of 127 additional sensors across PV and mechanical systems • Development of a unified data platform integrating existing BAS with new solar monitoring • Implementation of 5S in maintenance operations • Establishment of standardized maintenance procedures • Deployment of predictive maintenance algorithms for critical components World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1988 Figure 2 Implementation timeline and key milestones for the Phoenix case study 5. Results and Analysis Our implementation of integrated lean maintenance and smart monitoring across three case study sites yielded significant improvements in energy efficiency, maintenance costs, and system reliability. This section presents our key findings with supporting data. 5.1. Energy Efficiency Improvements The integration of lean maintenance with smart monitoring technologies delivered substantial efficiency enhancements across all implementation sites. As shown in Table 2, efficiency improvements ranged from 14.8% to 16.9%. Table 2 Energy Efficiency Improvements Following Implementation Site Location System Type Size Pre-Implementation Efficiency PostImplementation Efficiency Improvement A Phoenix, AZ Commercial Office 350 kW PV + HVAC 67.3% 82.1% 14.8% B Denver, CO Light Industrial 520 kW PV + Compressors 61.8% 78.5% 16.7% C Charlotte, NC University Campus 1.2 MW PV + District Cooling 72.4% 89.3% 16.9% These efficiency gains resulted from several key factors: • Reduction in unnecessary preventive maintenance activities that previously disrupted optimal system operation • More timely identification and resolution of performance issues • Real-time parameter optimization based on current conditions • Improved coordination between solar and mechanical systems • Reduced component wear through optimized operation World Journal of Advanced Research and Reviews, 2025, 26(03), 1981-1995 1989 "We discovered that approximately 34% of preventive maintenance activities performed under the previous regime were either premature or unnecessary," notes Dr. Elizabeth Chen, lead energy analyst for the project. "By transitioning to condition-based maintenance informed by real-time monitoring, we eliminated these disruptive interventions while actually improving system reliability." At the Charlotte campus, early detection of a 7% degradation in chiller efficiency prevented an estimated 23,000 kWh of wasted energy. At Phoenix, dynamic optimization of air handler fan speeds reduced energy consumption by 21% while maintaining occupant comfort. The Denver facility implemented intelligent energy storage by increasing compressed air production during peak solar generation periods (Reveles-Miranda, 2024.. Figure 3 Efficiency trend over 18-month implementation period across three sites The efficiency trends show continuous improvement throughout implementation, with significant gains during Phase 2 (Process Integration) and Phase 3 (Advanced Analytics). The improvement rate plateaued in later months, suggesting an approach toward theoretical maximum efficiency. 5.2. Maintenance Cost Reduction Our integrated approach delivered substantial reductions in maintenance costs, as detailed in Table 3. Table 3 Maintenance Cost Reductions by Category Cost Category Pre-Implementation ($/kW/year) Post-Implementation ($/kW/year) Reduction (%) Preventive Maintenance Labor 14.75 9.83 33.4% Corrective Maintenance Labor 18.32 12.91 29.5% Replacement Parts 22.67 15.18 33.0% Diagnostic Equipment 5.21 3.98 23.6% Specialized Services 11.33 9.87 12.9% Total 72.28 51.77 28.4%