Technologies Applied to Artificial Lighting in Indoor Agriculture: A Review

Abstract

Artificial lighting is essential in indoor agriculture, directly influencing plant growth and productivity. Optimizing its use requires advanced technologies that improve light management and adaptation to crop needs. This systematic review, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology, examines recent advancements in artificial lighting technologies, focusing on their applications, challenges, and future directions. A systematic search in Web of Science (WOS) and Scopus identified 70 relevant studies published between 2019 and 2024. The analysis highlights five major technology groups: (i) lighting control systems, with Light-Emitting Diodes (LEDs) as the dominant solution; (ii) Internet of Things (IoT) incorporating sensors, deep neural networks, Artificial Intelligence (AI), digital twins, and machine learning (ML) for real-time optimization, as well as communication technologies, enabling remote control and data-driven adjustments; (iii) simulation and modeling tools, refining lighting strategies to enhance plant responses and system performance; and (iv) complementary energy sources, improving lighting sustainability. IoT-driven automation has significantly improved artificial lighting efficiency, optimizing adaptation and plant-specific management. However, challenges such as system complexity, high energy demands, and scalability limitations persist. Future research should focus on refining IoT-driven adaptive lighting, improving sensor calibration for precise real-time adjustments, and developing cost-effective modular systems to enhance widespread adoption and optimize resource use.

1. Introduction

Indoor agriculture, also referred to as Controlled Environment Agriculture (CEA), is a modern farming technique that involves growing crops within a regulated environment [1,2]. This innovative approach leverages advanced technologies to precisely control essential growth factors, such as light, temperature, humidity, and nutrient availability [3]. By replicating and optimizing natural conditions, indoor agriculture allows for consistent plant growth regardless of external climatic or geographical constraints [3,4].

CEA systems incorporate a range of methodologies, including vertical farming, greenhouses, and plant factories, each utilizing distinct technological and environmental management strategies to optimize productivity and resource efficiency [2,5]. These systems offer the potential to revolutionize traditional agriculture by enhancing crop yields, reducing resource consumption, and enabling year-round production.

1.1. The Interest in Advancing Indoor Agriculture

The growing interest in enhancing indoor systems stems from their potential to address environmental, economic, and logistical challenges associated with traditional agricultural practices [6]. From an environmental perspective, integrating cutting-edge technologies such as structural and spatial systems (greenhouses and vertical farms), environmental control systems (climate control, light management, and atmospheric enhancement), advanced irrigation methods (aeroponic, hydroponic, and aquaponic), robotics and automation (farming robots and autonomous management systems), and IoT and sensing technologies (agricultural IoT sensors, equipment tracking, and environmental monitoring) creates a controlled environment that enables year-round crop production, while reducing the carbon footprint associated with transporting seasonal produce [7,8]. Each of these technologies has advantages that drive interest in their use and continuous improvement. For example, irrigation systems enhance water efficiency through recirculation and precision irrigation, significantly reducing water consumption compared to conventional methods [9,10,11,12,13]. Additionally, these systems minimize or eliminate pesticide use, promoting safer and more sustainable food production [14]. This continuous production capability also mitigates the economic volatility brought about by seasonal price fluctuations and the higher costs typically associated with off-season produce [9].

From a socio-economic perspective, indoor agriculture holds considerable promise in light of global trends such as population growth, urbanization, and the limited availability of arable land. By enabling crop cultivation within urban environments where traditional farming is less feasible, indoor agriculture contributes to food security and strengthens the resilience of local food systems [15,16,17]. Additionally, the adoption of technologies stimulates social and economic development by engaging professionals from diverse disciplines in the design, operation, and management of these systems. These advancements not only introduce modern technologies to current agricultural practices but also reduce the physical demands of labor through automation, thereby facilitating more efficient and sustainable farming operations in both urban and rural settings [4,9].

This framework of CEA thus represents a forward-thinking approach, positioning indoor agriculture as a key component in the future of global food systems.

1.2. Challenges of Indoor Agriculture

Despite its potential to address environmental, social, and economic challenges, indoor agriculture faces barriers that hinder its large-scale adoption [8,18,19]. One of the primary obstacles is the high initial investment required for facility construction, technology acquisition, land procurement, and the installation of advanced agronomic control systems. Additionally, specialized labor is essential for system operation and maintenance, increasing costs [9,19,20]. While cost-reduction strategies, such as comprehensive economic planning, advancements in cost-efficient technologies, and diversified funding models, can alleviate some financial burdens, the extended payback period discourages investors and slows widespread implementation [5,19,20].

Beyond financial constraints, high energy consumption poses another major challenge. Indoor agriculture relies heavily on artificial lighting, climate control, and irrigation systems to maintain optimal growing conditions, resulting in significant electricity usage and elevated operational costs [5,19,20,21]. Although energy-efficient technologies and renewable energy integration offer promising solutions, overall energy demands [5,20] and their associated environmental footprint remain substantial [19]. This creates a paradox, as indoor agriculture aims to reduce ecological impact but struggles with its resource-intensive nature.

In addition to economic and energy concerns, production limitations also restrict indoor agriculture’s scalability. Most systems are optimized for specific crops, such as leafy greens and herbs, limiting the variety of commercially viable produce. Furthermore, the yield per unit area in many indoor systems, particularly those in early development stages or lacking optimization, often falls short of the productivity levels achieved in traditional field farming, impacting economic feasibility [9,20]. Expanding production to a wider range of crops remains a challenge due to biological and economic constraints.

Finally, agronomic challenges, such as pest management [22] and nutrient delivery [23,24] under artificial conditions, remain areas of active research. Ensuring stable nutrient availability and effective pest control without traditional soil ecosystems requires innovative solutions. These factors underscore the complexity of developing sustainable, scalable, and economically viable indoor farming systems capable of competing with conventional agricultural methods.

1.3. Essential Conditions of Indoor Agriculture

To overcome the challenges identified in the previous section—such as high energy consumption, economic constraints, and crop limitations—indoor agriculture relies on several critical factors that enhance both efficiency and sustainability. These factors can be distilled into five essential conditions [25], each addressing a specific barrier to large-scale adoption and long-term viability.

1.3.1. Efficient Use of Space

Optimizing space is essential for enhancing productivity per unit of land [5,26]. Vertical farming, which involves stacking multiple layers of crops [27] and using movable growing beds that can be adjusted for various crops and growth stages [28], is standard practice to maximize both horizontal and vertical space. Such strategies enable high-density production, reducing the need for extensive land area and minimizing transportation-related emissions. Effective space utilization contributes to the overall sustainability of indoor agriculture by allowing more crops to be grown in smaller, controlled environments, particularly in urban settings where space is limited [29,30].

1.3.2. Light Optimization

Indoor systems rely on artificial lighting, such as light-emitting diode (LED) and high-pressure sodium (HPS) lights, to provide the appropriate light spectrum and intensity required for photosynthesis, phytochemical synthesis, and crop yield [31,32,33,34,35]. Advances in LED technology have improved energy efficiency and operational longevity, reducing energy consumption and operational costs [36,37,38,39,40]. Precise control of light exposure allows indoor systems to optimize plant growth, particularly in controlled environments where natural sunlight is limited or unavailable. Recent developments in tunable LED systems offer further opportunities to customize light conditions based on specific crop needs and growth stages, contributing to enhanced resource efficiency [28].

1.3.3. Water Conservation

Indoor agriculture is renowned for its effective water management by implementing advanced irrigation systems to ensure optimal water usage. Closed-loop hydroponic systems, for example, recirculate water and nutrients, significantly reducing water consumption compared to traditional agricultural methods [2]. Common techniques include the Nutrient Film Technique (NFT), where a thin layer of nutrient-rich water continuously flows over plant roots [41], and aeroponics, which involves suspending plant roots in the air and misting them with nutrient solutions to optimize oxygen exposure and nutrient uptake [13]. Deep Water Culture (DWC) submerges plant roots in nutrient solutions, while Ebb and Flow systems periodically flood and drain plant roots with nutrient solutions [42]. These systems enable precise control over water and nutrient delivery, reducing water usage by up to 90% compared to conventional agriculture, and are essential for promoting water conservation in indoor farming [43].

1.3.4. Automation and Mechanization

Automated systems manage key processes, such as planting, harvesting, irrigation, and environmental monitoring, reducing the reliance on manual labor and increasing the scalability of indoor systems [44]. Robotics and mechanized solutions further enhance precision in tasks such as planting and harvesting, while automated sensors continuously monitor environmental variables such as temperature, humidity, and nutrient levels, enabling real-time adjustments [44]. Integrating data-driven technologies and artificial intelligence allows for more informed decision-making and optimization of resource use, contributing to increased productivity and reduced labor costs [45].

1.3.5. Precise Control and Adjustment of Growing Parameters

Indoor systems allow for continuous monitoring and adjustment of both atmospheric and underground growth parameters. Atmospheric variables such as temperature, humidity, airflow, and CO₂ levels are regulated to create optimal conditions for plant growth [46,47], while underground factors like root zone temperature, water, nutrient concentration, and oxygen availability are tightly controlled to promote root health and nutrient absorption [48].

Among these essential conditions, light optimization is particularly crucial, as it directly influences photosynthesis, plant morphology, and overall productivity, while also being one of the most energy-intensive components of indoor agriculture. Given its fundamental role in determining efficiency and sustainability, the next section explores the significance of artificial lighting.

1.4. Artificial Light in Indoor Agriculture

Unlike natural sunlight, artificial lighting allows precise control over intensity, duration, and light quality, optimizing conditions for specific crop needs [49]. This precise management enhances productivity by ensuring that plants receive optimal conditions for critical processes such as biomass production, morphological development, and flowering [50]. Technological advancements, like tunable LED arrays and smart lighting systems with customizable outputs, have improved plant growth in CEA by allowing adjustments based on plant developmental stages and environmental conditions, maximizing light use while minimizing energy consumption [28,39,49,51,52,53].

Integrating sensors and renewable energy sources enhances system sustainability by ensuring optimal light delivery throughout the plant’s growth cycle [39]. In this context, understanding LED system performance is crucial. Efficiency refers to the amount of electrical energy converted into photosynthetically active photons, measured in micromoles per joule (µmol/J), while efficacy assesses how well the light supports plant growth by adjusting factors such as spectrum, intensity, and distribution [37,54]. Research has identified the most effective light spectra and timing strategies for various growth stages, improving photosynthesis and increasing crop yield. These LED innovations boost productivity and reduce energy costs in indoor farming systems [16,18,25,34,37,49,50].

This systematic review aims to investigate the technologies utilized in artificial lighting for indoor agriculture, providing a comprehensive analysis of advancements in this field. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology [55], this review identifies scientific publications from key databases to summarize the state of the art in CEA lighting technologies and related innovations and to evaluate their effectiveness in optimizing energy efficiency. The goal is to offer a foundation for future research by outlining research methodologies and proposing future directions to improve artificial lighting solutions in indoor agriculture.

2. Materials and Methods

2.1. Literature Search

This systematic review followed the PRISMA methodology, which provides a structured approach to identifying, screening, and assessing the quality of relevant studies. The PRISMA method ensures transparency and reproducibility by defining a research question, developing a search strategy, screening articles, and extracting data [55]. The central question of this review is: What technological applications in indoor agriculture enhance the efficient use of light optimization? Additionally, two sub-questions were developed to refine our focus:

• What lighting devices, as well as software, programs, or methods, are mostly used and manipulated in indoor agriculture?
• What are the trends and challenges in the technological applications of lighting in indoor agriculture?

The databases used for the literature search were Web of Science (WOS) and Scopus, selected for their extensive coverage of peer-reviewed studies across multiple disciplines. The search covered the period from 2019 to 2024, focusing on articles, reviews, and book chapters, with no restrictions on language or country of publication. Only Open Access publications were included to ensure accessibility.

For Web of Science, the search was performed across all databases and all collections using the topic field, which covers searches in the title, abstract, and indexing terms. In Scopus, the search was conducted using the article title, abstract, and keywords fields.

In alignment with the questions, this review focuses on technologies that enhance the efficient use of light in indoor agriculture. Consequently, it does not delve into establishing species-specific light requirements, as that falls under a different research scope. Addressing these needs would require a more tailored question and targeted search strategy based on specific crop types, each of which has distinct light needs.

2.2. Search Keywords

Before identifying the final set of keywords to answer the research questions, an initial exploration was carried out using several synonyms for the three main groups of terms: indoor agriculture, artificial lighting, and technologies.

The terms for the group indoor agriculture are used to describe crops grown in CEA; these terms include “Indoor crops”, “Indoor gardening”, “Indoor cultivation”, “Indoor horticulture”, “Indoor farming”, “Urban farming”, “Greenhouse cultivation”, “Greenhouse”, “Closed-system plant production”, “Plant factory systems”, “PFAL”, “Vertical farming”, “Urban agriculture”, “Hydroponic”, “Controlled environment agriculture”, “Plant cultivation in controlled environments”, and “Indoor plant cultivation”.

For the artificial lighting group, these terms refer to the application of artificial lighting in indoor plant cultivation, focusing on manipulating and controlling lighting systems, including various lighting technologies and aspects related to efficiency within the growth environment. Key terms include “Artificial lighting”, “Supplemental lighting”, “Artificial illumination”, “Photoperiod”, “Light-emitting diodes (LED) lighting”, “High-intensity discharge (HID) lighting”, “Fluorescent lighting”, “Light spectra”, “Light quality”, “Lighting efficacy”, “Light efficiency”, “Lighting intensity”, “Light distribution”, “Lighting uniformity”, “Lighting duration”, “Lighting technologies”, “Artificial light treatments” and “Artificial light sources”.

For technology, terms were selected to encompass automation and advanced technological approaches in agriculture. These include “Automation”, “Farm automation”, “Agricultural automation”, “Sensor”, “Smart farming”, “Precision agriculture”, “Precision farming”, “Automated systems”, “IoT”, “Machine Learning”, “Artificial intelligence” and “Robotic”.

This initial list of terms, along with all possible combinations, was employed in preliminary searches. However, it was observed that excessive use of specific terms frequently limited the search results significantly, occasionally resulting in no relevant records within the databases. Consequently, the preliminary searches using these specific terms or combinations proved overly restrictive in scope. This analysis informed the selection of the following keywords for the final search:

• Terms for indoor agriculture: Only the term “Agriculture” was used, as its combination with the other groups covered terms such as vertical farming, indoor agriculture, CEA, and plant factories, among others.
• Terms for artificial lighting: The key terms included “Light efficiency”, “Light spectra”, “Light-emitting diode lighting” combined with its acronym “LED”, “Light intensity”, and “Light”.
• Terms for technology: The terms used were “Artificial intelligence” combined with its acronym “AI”, “Internet of Things” combined with “IoT”, “Sensor”, and “Automation”.

The search was conducted by combining one keyword from each group. These combinations were constructed using the boolean operator “AND” between groups, while terms with acronyms were placed in parentheses and separated by “OR”. This process resulted in 20 different search combinations for each database to ensure comprehensive coverage of the relevant literature (

Table 1. Search queries and document counts in Scopus Database.

AND		Artificial Lighting Terms					AND
		“Light Efficiency”	“Light Spectra”	“Light-Emitting Diode” OR “LED”	“Light Intensity”	“Light”
Technology terms	“Artificial intelligence” OR “AI”	126	54	1666	70	842	“Agriculture”	Indoor agriculture term
	“Internet of Things” OR “IoT”	75	28	493	55	373
	“Sensor”	243	152	1857	214	1619
	“Automation”	40	4	235	14	159
Total results		8319

and

Table 2. Search Queries and Document Counts in WOS Database.

AND		Artificial Lighting Terms					AND
		“Light Efficiency”	“Light Spectra”	“Light-Emitting Diode” OR “LED”	“Light Intensity”	“Light”
Technology terms	“Artificial intelligence” OR “AI”	34	7	2	10	140	“Agriculture”	Indoor agriculture term
	“Internet of Things” OR “IoT”	52	11	16	68	243
	“Sensor”	97	58	34	106	585
	“Automation”	27	4	4	20	108
Total results		1626

).

2.3. Screening and Selection Stage

The initial search retrieved a total of 9945 documents: 8319 from Scopus and 1626 from WOS. After removing 3621 duplicates, 6324 unique articles remained. The selection process began by excluding 5845 documents based on their titles, as they did not align with the objectives of this review—many were unrelated to indoor agriculture or did not specifically address artificial lighting. Additionally, 17 duplicate titles were removed, leaving 462 articles for further evaluation. Next, articles were reviewed in three stages: first by analyzing keywords, then by screening abstracts, and finally by assessing full texts. This led to the exclusion of 392 articles that did not meet the study’s research objectives or methodological criteria. In the end, 70 references were selected for inclusion in this systematic review (Figure 1).

3. Results

3.1. Literature Analysis

A total of 70 articles were identified in this systematic review, with 4 articles published in 2019, 3 in 2020, 13 in 2021, 16 in 2022, 15 in 2023, and 19 articles published up to September 2024. Regarding document types, sixty-two were classified as original research articles, six as review articles, and two as book chapters. All articles were published in English.

The largest contribution came from Asia, with forty-three articles, including China (fourteen), India (eight), Malaysia (six), Taiwan (four), South Korea (three), Thailand (two), Pakistan (two), Iran (two), Japan (one), and the United Arab Emirates (one). Europe contributed fifteen articles, with Italy (four), Greece (two), the Netherlands (one), Germany (one), Poland (one), Portugal (one), Romania (one), Russia (one), Spain (one), Hungary (1), and Turkey (1). The Americas accounted for 11 articles, with the United States (seven), Canada (three), and Colombia (one). From Africa, one article was identified: Nigeria (one). No articles were contributed from Oceania. The geographical distribution of these publications was analyzed based on the primary affiliation of the corresponding author. This classification provides a clear view of the global research landscape, with a significant concentration of studies from Asia, Europe, and the Americas (Figure 2).

Across all continents, research experiments were conducted in both indoor environments and hybrid indoor–outdoor setups, focusing on artificial light control and supplementing natural light. However, research involving simulations was consistently applied across studies from Asia, Europe, and America.

In Asia, research primarily aimed to control plant growth in response to light conditions to optimize yield or adjust light based on plant requirements throughout the growth cycle. European research focused similarly on plant growth control, with an emphasis on improved energy efficiency. American studies also concentrated on controlling plant growth and light conditions for optimal yield, while emphasizing energy efficiency and cost reduction for indoor agricultural systems. In contrast, the single study from Africa centered more on methods for measuring light intensity.

To contextualize the focus of the 70 reviewed articles, a term frequency analysis was applied, extracting the 30 most frequently occurring words using Python 3.12 programming language to process the texts and filter the core terms in the literature. This approach was chosen to reinforce the alignment of the articles with the research objectives and questions. The top 10 terms “Data”, “Light”, “System”, “Plant”, “Energy”, “Control”, “Sensor”, “Model”, “Growth”, and “Greenhouse”, highlight the emphasis on data-driven methods, artificial lighting optimization, energy efficiency, and plant growth monitoring in indoor agriculture (Figure 3). These terms support the research questions and provide a basis for evaluating current trends of technological applications in light optimization for indoor agriculture.

3.2. Technologies Used in Indoor Artificial Lighting of Plants

Based on the methods and objectives outlined in the seventy articles included in this systematic review, four key technology groups have been identified as relevant for optimizing artificial lighting in indoor agriculture. Each group encompasses various approaches and tools aimed at enhancing lighting efficiency, reducing energy consumption, and improving plant growth conditions. While some technologies are frequently referenced across multiple articles, their specific applications vary depending on agricultural requirements.

Table 3. Overview technologies used in indoor artificial lighting of plants.

Technologies	Application	Source
Lighting control Light-emitting diode (LED) Systems with adjustable spectra, including Ultraviolet, Visible spectra, and Near-Infrared. Pulse Width Modulation (PMW) LED.	Spectrum management; Photoperiod configuration; Chromatic composition; Intensity control; Supplemental lighting; Adjustable PAR (Photosynthetically Active Radiation); PPFD (Photosynthetic Photon Flux Density) control; Automation and remote control; Uniform light distribution, moving and adaptive grow lights;	[18,47,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92]
Interconnected network Internet of Things (IoT) Sensors e.g., AS7265x IoT Sensor, VTB8440BH Photodiode Sensor, dual color detectable PAR sensor (combining SiPD and ripple-free DBPF), YQZBML Mod-lights, TCS32xx family of color sensors, spectroradiometers. Deep neural network e.g., Raspberry Pi, Arduino, TensorFlow. Artificial intelligence (AI) e.g., Fuzzy logic, Image analysis, AI techniques. Cyber-Physical Systems e.g., Digital twin. Machine learning e.g., Algorithms, MNIST dataset, Proportional-Integral-Derivative (PID), Fuzzy logic (FL), neural network, deep learning, hybrid control algorithms, Model predictive control (MPC), Sliding window-based support vector regression (SW-SVR), Space Colonization Algorithm (SCA), ANN, SVM, and SSD. Communication technology Wireless Sensor Networks (WSNs), e.g., Blynk cloud server, LoRa, LoRaWAN, Message Queuing Telemetry Transport (MQTT), NB-IoT, New Generation Service Interfaces for Linked Data (NGSI LD), ZigBee, 6LoWPAN, Wi-Fi, near field communication, ZigBee, Bluetooth low energy, mobile-phone technologies, Dash-7, Ethernet.	Remote control of parameters; Integration of cloud data over a mobile phone application; High-precision digital sensors for light intensity monitoring and measurement; Real-time monitoring and optimization of PAR and PPFD; Integration of cloud computing capabilities, personal digital assistants, control devices, and operational terminals; Autonomous adjustments or operation of light control systems without external sources; Automatic lighting adjustment for crop growth and health control.	[18,39,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,87,88,89,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117]
Simulations e.g., U-chord curvature, Qualnet simulator, prediction models.	Optimizing light angles for maximum photon absorption; Simulating light spectrum effects on crop growth. Simulation of luminous and energy efficiency;	[28,47,57,58,66,100,101,107,118,119,120]
Complementary energy sources e.g., photovoltaic (PV) systems, double-networking PAA-RGO-PANI hydrogel, and greenhouses with complementary solar energy.	Balancing natural and artificial light sources for energy efficiency; Reduction in operational costs and grid dependency.	[56,60,86,93,95,104,106,116,121]

provides an overview of these technologies and their primary applications.

The technologies implemented in artificial lighting systems for indoor agriculture are broad, diverse, and often tailored to the system’s complexity, specific application objectives, and available economic resources. Table 1 summarizes this extensive landscape, categorizing the findings into four groups: (1) lighting control systems, (2) IoT integration and communication technologies, (3) simulations, and (4) complementary energy sources.

The first group, lighting control systems, primarily focuses on using LED technologies, which dominate artificial lighting in plant cultivation due to their adaptability and efficiency. While other lamps are occasionally employed, their use is negligible compared to LEDs. LEDs are predominantly applied to achieve optimal light quality, which refers to the spectrum and intensity that best meet the specific requirements of a given species or agricultural purpose.

The second group—IoT integration and communication technologies—functions as a cohesive framework enabling artificial lighting systems’ systematic and autonomous operation. IoT, defined as the interconnection of devices that collect, process, and share data, incorporates sensors that gather raw lighting data. These data are transmitted via wireless sensor networks, which serve as a critical communication bridge, ensuring seamless delivery to centralized systems or cloud platforms for further analysis. As advanced analytical tools, machine learning and deep neural networks process these datasets to identify patterns, predict optimal lighting parameters, and generate actionable insights. AI, acting as the overarching system, integrates these analyses to make automated and precise adjustments to lighting conditions. Cyber-physical systems validate these AI-driven decisions by simulating their impact on the physical environment, ensuring accuracy and effectiveness before implementation. This interconnected framework—powered by IoT, communication technologies, and advanced analytical systems—optimizes lighting operations, enhancing energy efficiency and crop productivity while minimizing human intervention.

The third group, simulations, includes prediction models and tools for analyzing the effects of artificial lighting manipulation on agricultural productivity. Finally, the fourth group, complementary energy sources, focuses on balancing natural and artificial light to improve energy efficiency and reduce operational costs.

4. Discussion

4.1. Purposes and Applications of Indoor Artificial Lighting Technology

The unifying purpose of the 70 reviewed articles is to optimize light efficiency to maximize agricultural production, ensure food security, and create adaptive lighting environments tailored to crops’ physiological and developmental needs. This is achieved through advanced technological solutions, including automation systems, real-time monitoring, and precision control mechanisms, which enhance the quality and efficiency of artificial lighting across diverse cultivation scenarios. Additionally, these technologies integrate renewable energy sources and complementary systems to reduce operational costs and improve sustainability in large-scale indoor farming operations.

As summarized in Table 3, advancements in light optimization have revolutionized indoor agriculture, providing precise and dynamic control over environmental factors. Photobiology, the study of interactions between light and living organisms, forms the theoretical foundation for these innovations [122]. Core principles such as spectral composition, light intensity, photoperiod configuration, PPFD, and spatial light distribution are critical factors in maximizing the efficiency of artificial lighting systems in controlled environments, as supported by [28,57,63,94,115].

The visible spectrum, particularly the red and blue wavelengths within the PAR range (400–700 nm), is pivotal in driving photosynthesis and regulating photoreceptor regulation [18,47,84,91,97]. Red and blue light combinations have demonstrated significant benefits in a variety of crops, including rocket [79], cucumber [58], water spinach [88], radish [92], and lettuce [89]. Monochromatic treatments, such as red-light application in mint [68] or higher red-light ratios in bok choy [76] and sunflower [81], have been shown to enhance growth performance and alter phytochemical profiles. Additionally, integrating white or supplemental natural light has been explored to optimize plant development across multiple species.

Beyond the PAR spectrum, research on ultraviolet (UV) and infrared (IR) wavelengths highlights their contributions to plant physiology and secondary metabolite production [39]. For instance, UV light enhances secondary metabolite synthesis in strawberries [77] and triggers specific physiological responses in other crops [74,83]. Similarly, IR light has been utilized to manipulate flowering dynamics and elongation in species like bok choy [80]. These strategies yield notable improvements, including increased photosynthetic efficiency [58,90,101,113,119], greater biomass production [80,85,120], higher yields [59,112], enhanced growth dynamics [76,88,109], and superior pre- and post-harvest quality [83,118].

The optimization of PPFD has also emerged as a key technological application in indoor agriculture. Precise measurement and regulation of light availability enhance energy efficiency and crop growth [82,95,101,102,119]. Methods such as the U-chord curvature approach have been applied to determine optimal PPFD values, as demonstrated in studies with tomato seedlings [120]. Tailored adjustments based on crop-specific requirements have further refined these applications, with findings indicating optimal PPFD levels of 150 µmol·m⁻²·s⁻¹ for tea plants [67], 243 µmol·m⁻¹·s⁻¹ for tomatoes [120], and approximately 300 µmol·m⁻²·s⁻¹ for bok choy [80], each maximizing photosynthetic efficiency and biomass production.

Smart lighting technologies have advanced through the integration of innovative solutions, particularly those based on IoT, which encompass sensors, automated control devices, and advanced data communication infrastructure. These innovations enable real-time monitoring, remote control, and autonomous adjustments, optimizing light utilization while minimizing human intervention [18]. Numerous studies have demonstrated the effectiveness of IoT-driven smart indoor farms, leveraging cutting-edge digital and physical technologies to automate lighting control and optimize energy consumption [47,59,112]. For example, systems can dynamically adjust light intensity to meet species-specific needs, thereby reducing energy costs while improving productivity [113], or adapt photoperiod settings to regulate light duration and intensity for maximizing crop yield. In mint, for instance, photoperiods ranging from 8 to 16 h have been shown to increase extract production and improve growth conditions [68]. Similarly, pulsed lighting strategies in bok choy have enhanced leaf area, biomass, and chlorophyll content while maintaining energy efficiency [76].

The appropriate selection of intelligent architectures is crucial for achieving maximum efficiency, depending on the specific application and operational requirements. First, robust sensor and device integration must ensure seamless connectivity and data transmission through advanced communication protocols, guaranteeing efficient system operation. Second, data processing systems leveraging Big Data analytics, machine learning, and cloud computing play a pivotal role in analyzing environmental conditions and optimizing resources such as light and energy. These technologies enable autonomous management, reducing the need for human intervention while enhancing the precision of light control strategies. Third, predictive analytics and real-time alert systems facilitate proactive decision-making, ensuring rapid responses to environmental fluctuations and system failures, as demonstrated in studies focused on intelligent monitoring and management systems [18].

Practical implementations of these technologies include interconnected master–slave device networks that regulate light conditions in growth chambers while enabling remote monitoring through wireless communication [39]. Additionally, advanced algorithms and deep learning techniques have significantly improved the efficiency of smart lighting systems, allowing automated adjustments based on real-time crop requirements. As highlighted in studies such as [78,115], real-time environmental monitoring, combined with intelligent sensors, ensures that artificial light is deployed only when necessary, thereby reducing excessive energy consumption and enhancing sustainability.

The advent of remote control platforms has improved the flexibility and efficiency of indoor agriculture. Modern applications allow farmers to access, monitor, and control their farming environments via cloud-based mobile applications, eliminating the need for constant physical supervision. For example, GreenLab, a small-scale smart greenhouse system, integrates mobile connectivity with real-time data visualization, enabling users to remotely adjust temperature, humidity, and light conditions [106]. Similarly, Ref. [87] connected sensors and actuators to cloud-based platforms, allowing for precise remote lighting management while supporting real-time data storage and analytics.

Another significant advancement in remote monitoring systems is the development of real-time alert notification technologies, such as those implemented in [69]. These systems employ GSM and Arduino-based technologies to send SMS alerts to farmers regarding lighting status changes, ensuring rapid responses to system failures. By minimizing manual intervention, these solutions enhance operational efficiency and reliability in controlled environments.

Beyond remote monitoring, renewable energy integration is increasingly being implemented to mitigate the costs associated with artificial lighting, with a strong emphasis on photovoltaic-powered sensors. Studies such as [64,93] have demonstrated the feasibility of solar-powered monitoring devices. Additionally, hybrid lighting systems that combine natural and artificial light have gained attention, as evidenced by research on adaptable lighting frameworks [60]. One notable example is the use of automated roll-up shading systems in solar greenhouses, which optimize natural light utilization and reduce artificial lighting needs, leading to a reported 29% improvement in energy efficiency and lower electricity consumption in supplemental lighting systems [121].

Simulation-based strategies further contribute to optimizing lighting efficiency. Predictive control systems, supported by advanced simulations, allow for the seamless integration of artificial light management. Transient models are used to predict energy losses and gains, enabling researchers to identify the most efficient designs for indoor farming structures. Some studies, such as [61], have focused on digital twin modeling to simulate lighting scenarios and analyze energy consumption patterns in indoor agriculture. Similarly, wireless sensor networks and optimization algorithms have been applied to regulate artificial lighting and ventilation, ensuring real-time adjustments for maximum efficiency [112].

Other simulation methodologies focus on optimizing light distribution within controlled environments. Photon capture mathematical models have enhanced PAR efficiency, significantly improving crop yield and energy conservation. As demonstrated in [28], optimizing light penetration within plant canopies based on different indoor structural designs has shown measurable improvements in overall agricultural output.

4.2. Primary Devices and Software, Program or Method System Used in Lighting Indoor Agriculture

LEDs have emerged as the predominant lighting technology in indoor agriculture due to their high energy efficiency, minimal heat production, tunable spectral composition, and long lifespan [63,75,81,88,89]. While LEDs are the most employed lighting systems, studies continue to explore alternative illumination sources, such as fluorescent lamps and emerging materials designed to enhance LED efficiency and spectral performance. For example, studies have investigated the impact of fluorescent light on plant growth parameters, with some findings suggesting improved leaf expansion under fluorescent conditions compared to certain LED treatments. However, fluorescent lighting generally has lower energy efficiency and limited spectral control, making it less suitable for modern precision agriculture systems [84]. Additionally, recent research introduced Mn⁴⁺-doped Sr₂GdTaO₆ (SGTO:Mn⁴⁺), a thermally stable deep-red phosphor synthesized via a high-temperature solid-state reaction, which shows promising potential for enhancing LED-based plant lighting [90].

In IoT-driven CEA, Arduino and Raspberry Pi are among the most widely used hardware platforms in Asia and Europe, serving as core technologies for automation, robotics, and IoT. Although they share some common ground, their roles in artificial lighting applications differ significantly in terms of computational capabilities, power efficiency, and real-time control. A deeper understanding of these differences is essential for selecting the most suitable platform and developing optimized IoT-based systems.

Arduino is particularly suited for applications requiring precise sensor and actuator control with low power consumption, making it ideal for real-time embedded systems that do not rely on an operating system. Its effectiveness in artificial lighting control stems from its ability to directly interface with sensors and actuators without unnecessary computational overhead. For instance, Arduino-based IoT systems dynamically regulate lighting activation based on environmental conditions, acting as photoperiod regulators that minimize energy waste while maximizing efficiency [68,77]. Similarly, Arduino UNO with relay modules enables precise modulation of the LED spectrum and intensity, tailoring lighting conditions to support plant growth and improve agricultural yield [92]. By integrating Light Dependent Resistor (LDR) sensors—widely referenced in Asian studies—these systems enhance automation by activating lighting only when necessary, according to ambient light levels, further optimizing energy consumption [71]. Additionally, as previously mentioned, Arduino allows remote monitoring and alerts, reinforcing its role as a versatile and efficient control system [105]. Its ability to provide reliable, low-cost, and energy-efficient control makes it particularly advantageous for artificial lighting applications where simplicity and precision are paramount.

On the other hand, Raspberry Pi is designed for multitasking and high processing power, making it ideal for applications requiring internet connectivity, database management, and AI-driven decision-making. Its ability to execute multiple tasks simultaneously enables advanced automation strategies, particularly in artificial lighting systems that rely on adaptive, data-driven optimization. One notable application is its use in adjusting LED light intensity via pulse-width modulation (PWM), dynamically regulating power delivery based on sensor inputs to optimize energy efficiency and maintain ideal illumination levels [60]. Beyond lighting control, Raspberry Pi also integrates sensor networks to regulate environmental conditions such as temperature and humidity, ensuring a stable growth environment [47].

A key component in this automation is Blynk, an IoT platform that facilitates seamless cloud-based monitoring and control. Its widespread adoption, particularly in Asia, is attributed to its efficient integration with microcontrollers and its capability to centralize data from multiple sensors. Through Blynk, Raspberry Pi enables remote management of lighting and climate parameters while storing environmental data for further analysis, supporting predictive control strategies [71,81]. Additionally, Blynk’s cloud storage and IoT connectivity enhance Raspberry Pi’s functionality, allowing real-time system adjustments and automated decision-making based on live sensor feedback [88,89].

While both platforms offer unique advantages, their limitations must be considered in the context of artificial lighting automation [94]. Arduino’s simplicity and efficiency come with limited multitasking capabilities, restricting its ability to process complex data. Conversely, Raspberry Pi, though more powerful, relies on an operating system that introduces latency, making it less suitable for real-time actuator control. However, integrating both in a hybrid IoT architecture provides a robust solution: Raspberry Pi handles high-level tasks such as image processing, data analysis, and network communication, while Arduino ensures low-latency execution of actuator commands for precise lighting adjustments. This combination enhances adaptability and precision, optimizing lighting conditions dynamically [65].

The integration of AI and machine learning (ML) within IoT-driven agricultural systems has significantly improved the automation of lighting regulation. However, selecting the most effective algorithm remains a challenge, as models vary in accuracy, efficiency, and computational demand depending on the environmental parameters being controlled [63].

ML has demonstrated superior capability in handling non-linear relationships in agricultural applications, particularly in processing sensor data for precision farming [78]. However, traditional approaches such as linear regression (LR) and non-linear regression (NLR) present limitations in modeling complex variable interactions and are highly sensitive to data reliability [113]. To address these constraints, more advanced models have been explored, including Random Forest Regression (RFR), Support Vector Regression (SVR), and Backpropagation Neural Networks (BPNNs). These algorithms have shown superior performance in predicting photosynthetic rate (Pn), a key factor in optimizing light exposure. For example, the RFR algorithm outperformed others in predicting cucumber Pn under varying light intensities [113], while SVR exhibited higher accuracy than RFs, BPNNs, and NLR in Pn modeling based on light intensity (PPFD), demonstrating its potential for precision agriculture [119]. Additionally, hybrid approaches combining SVR with Particle Swarm Optimization (PSO) have been employed to enhance Pn accuracy. The Normal Boundary Intersection (NBI) algorithm has been used to determine optimal light control points, preventing energy waste while ensuring optimal plant growth [101].

In parallel, PID controllers, combined with clustering algorithms such as K-Means++ and Furthest Priority K-Means (FPKM), dynamically adjust lighting intensity based on model outputs, improving both energy efficiency and system stability. These approaches enable real-time monitoring and adaptive light regulation in greenhouses, ensuring precise environmental control while optimizing agricultural productivity. Moreover, they enhance data management by reducing storage demands and computational iterations, thereby improving overall system performance [18,78,115].

Beyond ML-based optimization, AI-driven techniques have been applied to optical design in controlled environments. Genetic Algorithms (GAs), integrated with digital twin simulations and ray-tracing methods, have been used to optimize LED configurations—such as beam aperture, source tube size, and power distribution—to maximize energy absorption and improve light propagation [57]. Additionally, Radial Basis Function (RBF) neural networks, optimized with Quantum Genetic Algorithms (QGAs), have demonstrated high accuracy in predicting Pn, providing a reliable foundation for automated light control in smart agriculture [58].

IoT-driven communication technologies are fundamental for smart agriculture. The most used protocols include Wi-Fi, Bluetooth, ZigBee, LoRaWAN, Sigfox, RFID, and cellular networks (2G, 3G, 4G, and 5G) [18,112]. While Wi-Fi and Bluetooth offer high-speed data transfer and easy integration in small-scale applications, ZigBee is more energy-efficient and suitable for large-scale sensor networks. Long-range, low-power communication protocols such as LoRaWAN and Sigfox provide extensive coverage but at the expense of data transfer speed. Cellular networks, particularly 5G, enable real-time, high-bandwidth communication but are costly and require substantial infrastructure investment.

4.3. Challenges, Future Research, and Trends in Artificial Lighting for Indoor

The optimization of artificial lighting faces three main challenges: high energy consumption, system complexity, and scalability. Although advancements in LED technology, IoT, communication technology, simulations, and renewable energy integration have improved efficiency, artificial lighting remains a major operational cost. Future efforts should prioritize self-regulating lighting systems that dynamically adjust to plant needs while minimizing energy demand, further leveraging renewable energy sources and predictive control mechanisms to enhance sustainability [61].

A second challenge lies in the integration of IoT-driven automation. Current models struggle with real-time adaptability due to variations in sensor accuracy and environmental conditions [47]. Future research should focus on machine learning models with continuous learning capabilities, allowing dynamic adjustments in spectral composition, intensity, and photoperiods. Additionally, standardized sensor calibration and improved data processing algorithms will enhance the reliability of automated lighting control [96].

Scalability remains another barrier, particularly for small and medium-sized farms, where the high initial investment and technical complexity limit adoption. Future efforts should prioritize cost-effective, modular lighting solutions, leveraging low-power computing, edge IoT, and improved communication protocols to enable affordable, decentralized control systems [59,93].

Emerging research trends indicate a shift toward digital twin simulations, adaptive sensor networks, and biofeedback-driven lighting adjustments. These advancements aim to reduce trial-and-error costs and improve system efficiency. Furthermore, sustainability-driven innovations are exploring alternative light sources, hybrid natural–artificial lighting strategies, and IoT-powered energy management to optimize resource use while maintaining high crop productivity.

5. Conclusions

This systematic review underscores the critical role of advanced technologies in optimizing artificial lighting systems for indoor agriculture. The analysis of 70 scientific publications reveals a steady increase in research over the past five years, with contributions predominantly from Europe, North America, and Asia.

LED systems remain the most widely employed technology due to their high energy efficiency, spectral tunability, and adaptability. However, their integration with IoT systems has significantly enhanced their efficacy. These advancements enable precise environmental control, real-time monitoring, and adaptive lighting adjustments that optimize photosynthetic efficiency while minimizing energy consumption and operational costs.

Simulation models have emerged as a key tool for evaluating lighting configurations before implementation. These models allow for the assessment of various environmental conditions, improving decision-making, reducing trial-and-error costs, and enhancing overall system efficiency. Despite inherent complexities, predictive modeling contributes to structured lighting management strategies that maximize resource utilization.

The review also highlights the increasing reliance on IoT-driven automation for lighting control. By integrating wireless sensor networks, deep learning algorithms, and real-time data analytics, modern systems dynamically adjust lighting parameters to align with plant growth requirements. These smart systems improve energy efficiency and streamline operations, reducing human intervention.

Despite these technological advancements, challenges remain. High energy demands, system complexity, and scalability limitations hinder widespread adoption. Future research should enhance machine learning models for real-time adaptability, improve sensor calibration, and develop cost-effective, modular lighting solutions. Additionally, integrating renewable energy sources and hybrid lighting strategies will be crucial for long-term sustainability.

Overall, no single approach is universally optimal. The reviewed literature highlights the benefits of integrating technologies from different domains—such as lighting control systems, IoT-based automation, simulations, and complementary energy sources—to enhance productivity, automation, and sustainability in indoor agriculture.