As carbon neutrality become an increasingly urgent issue, the challenge of managing increasingly complex and renewable-heavy power systems remains. Work in the International Journal of Reasoning-based Intelligent Systems looks at power system optimization and presents a new approach to reducing both emissions and operating costs using artificial intelligence, AI.

The context for this innovation is the dual carbon goal, a policy framework adopted by many countries aiming to reach carbon neutrality sooner, rather than later. As electricity grids evolve from centralized, fossil-fuel-based structures to more decentralized systems dominated by intermittent renewable sources like wind and solar, they become harder to control in real-time. Grid operators must now balance not only power supply and demand, but also the economic cost of power generation and its environmental impact.

The researchers have used a refined version of a computational technique known as Particle Swarm Optimization (PSO). Originally inspired by the coordinated movement of animals, PSO seeks out optimal solutions from complex systems by simulating swarming behaviour. The researchers have extended PSO with a more advanced variant called Multi-Strategy Adaptive PSO (MAPSO), which incorporates a dynamic "reward mechanism" to improve how the AI searches for optimal outcomes. The researchers have then taken this yet another step forward, with MOMAPSO (Multi-Objective MAPSO). It operates within a dual-objective framework that simultaneously minimizes fuel costs and pollutant emissions, while respecting essential operational constraints such as power balance and generation limits.

Traditional mathematical optimization methods, while theoretically accurate, often struggle under the weight of modern grid complexity. They tend to be computationally demanding and less effective at navigating trade-offs between competing objectives. Heuristic algorithms like PSO and the extended version MAPSO, by contrast, offer a more flexible and efficient alternative.

Yao, L. (2025) 'Optimisation of operation algorithms based on artificial intelligence in power system control', Int. J. Reasoning-based Intelligent Systems, Vol. 17, No. 9, pp.34–43.
DOI: 10.1504/IJRIS.2025.147656

In an obvious departure from conventional wisdom regarding tourism, research in the International Journal of Business Innovation and Research shows that domestic travellers in Indonesia are prioritising safety over traditional markers of quality, such as service excellence. The study, based on data from 300 tourists across Jakarta, Banten, and West Java, suggests that the post-pandemic traveller is guided less by indulgence and more by psychological comfort and trust in a destination.

The researchers draw on the Theory of Planned Behaviour, a social psychology model commonly used to explain how individuals form intentions based on attitudes, social norms, and perceived control. They then examined how a tourist's perception of a destination and the risks associated with a visit to that place are now shaping travel decisions. The findings indicate a fundamental shift in the Indonesian tourism landscape: travellers are no longer primarily attracted by high-end service offerings. Instead, they favourable a strong destination image that encompasses traveller safety, emotional appeal, cultural assets, and perceived value.

Where such positives might encourage travel to a particular tourist spot. Any perceived risk to personal safety and wellbeing acts as a strong deterrent. Fears regarding health and safety risks, as well as concerns about financial loss or the reliability of infrastructure and local services feed into this deterrent, the research shows. Even destinations with traditionally high service standards are seeing waning interest if they cannot address these apparently new psychological barriers.

One of the most notable demographic findings from the study is the emergence of middle-class women aged 30 to 40 as the most influential segment of the domestic market. For this group, affordability, consistency, and emotional well-being rank higher than novelty or luxury. Their preferences are shaping a broader trend toward more cautious, purpose-driven travel.

Ultimately, the research has implications beyond Indonesia if the same personal risks assessments are being applied to other destinations in South East Asia. For tourism operators and policymakers across the globe, these insights suggest the industry needs to recalibrate its marketing and operational strategies to take into account changes in the attitudes of today's tourists.

Ambarwati, M.F.L., So, I.G., Abdinagoro, S.B. and Pradipto, Y.D. (2025) 'Recalibrating the compass: what truly matters to travellers after uncertainty', Int. J. Business Innovation and Research, Vol. 37, No. 6, pp.1–21.
DOI: 10.1504/IJBIR.2025.147657

Research in the International Journal of Vehicle Performance discusses a new integrated sensing and positioning system that can improve how autonomous locomotives navigate underground mining tunnels. This is one of the most challenging environments for vehicle automation, given the potential for hazards. The team has combined advanced laser-based mapping with wireless positioning and demonstrates that their system promises safer and more reliable navigation in settings where traditional sensing technologies often fail.

Underground mining poses distinct problems for autonomous vehicles. Tunnels are narrow, poorly lit, and often filled with dust and smoke, conditions that render conventional cameras almost useless and reduce the effectiveness of radar. This study addresses these challenges with a hybrid approach that pairs Light Detection and Ranging (LiDAR) sensors with Ultra Wide Band (UWB) wireless positioning.

LiDAR works by emitting laser pulses and measuring the time it takes for the light to return after bouncing off nearby objects. Unlike cameras, LiDAR does not depend on ambient light and can generate detailed 3D maps even in visually obscured environments. But the raw data it produces is often noisy and difficult to interpret, especially in cluttered or uneven spaces like those found in mining tunnels.

To refine this data, the researchers employed a technique known as multi-iteration plane fitting, which systematically isolates the ground surface from the rest of the environment. This step is essential for identifying true obstacles rather than confusing irregularities in the terrain for actual hazards. Once the ground is mapped, remaining data points, representing non-ground objects, are grouped into clusters.

Clustering in irregular environments presents its own difficulties. Traditional algorithms can misinterpret the scene, either by merging separate objects into one or splitting a single object into several false positives. The research team addressed this by analysing the scan line distribution from the LiDAR sensors, that is, the geometric pattern in which the laser sweeps across the environment. This allowed them to group points that belong to the same physical object more effectively. This approach not only improves obstacle recognition accuracy but also speeds up data processing by more than a fifth, which is important for systems operating in real time.

Wang, H., Wang, Y. and Shen, Y. (2025) 'Identification of intrusion obstacles for underground locomotives based on the fusion of LiDAR and wireless positioning technology', Int. J. Vehicle Performance, Vol. 11, No. 3, pp.253–277.
DOI: 10.1504/IJVP.2025.147693

As artificial intelligence (AI) continues to transform how images and video are created, concerns about the manipulation of digital content have grown. Research in the International Journal of Forensic Engineering discusses how AI can be used to fight back and find the fakes.

Deepfakes, videos or images that have been manipulated using AI to depict events that never occurred and words that were never uttered, pose a growing threat to public discourse, politics, criminal investigations, and scientific integrity. These convincing forgeries are typically generated by a form of AI known as Generative Adversarial Networks (GANs). In a GAN, two neural networks compete: one creates realistic fake images, while the other attempts to identify the deception. As they evolve, so the system becomes adept at producing forgeries that are difficult even for experts to distinguish from authentic content.

Such technologies have not only led to blurred lines between real and artificial but also raised critical questions for journalism, law enforcement, and scientific publishing, where the authenticity of visual evidence can have serious consequences. As the manipulation of media becomes more sophisticated, traditional detection methods, such as spotting anomalies in shadows or unnatural features, have proven inadequate.

The work in IJFE has repurposed a relatively simple machine-learning model to detect deepfakes with high accuracy. The researchers tested three convolutional neural network (CNN) architectures, ResNet-50, AlexNet, and LeNet-5, to determine which was most effective in distinguishing genuine images from AI-generated fakes.

CNNs are particularly good at analysing visual data. They function by identifying patterns, textures, and other features that may not be visible to the human eye. Although newer models like ResNet-50 are often assumed to be more powerful due to their increased sophistication, the researchers found that LeNet-5, a comparatively basic model developed in the 1990s, initially performed best.

The team was able to enhance this CNN through architectural fine-tuning and parameter optimisation, so that their modified version of the LeNet-5 model could reach almost 96 percent accuracy. It also had a demonstrable Area Under Curve (AUC) score of almost 99 percent, meaning that statistically it performed well across a range of test scenarios. It is worth noting that this system does not rely on prior knowledge about an image's content nor the embedded metadata.

Of course, as researchers are developing the tools to detect the fakes, so those who develop the tools to create them in the first place will respond with increasing sophistication to outwit the detectors.

Pawade, V.S. and Mathur, S. (2025) 'Comparative study of CNN models for detecting altered and manipulated images', Int. J. Forensic Engineering, Vol. 5, No. 3, pp.216–227.
DOI: 10.1504/IJFE.2025.147560

A newly developed performance evaluation system can assess energy storage power plants (ESPPs) in terms of investment decisions and public policy, as well as their place in sustainable energy infrastructure. The research is discussed in the International Journal of Power and Energy Conversion.

A longstanding and well-known bottleneck in sustainable energy integration is how to reliably and comprehensively evaluate the complex trade-offs involved in operating electrochemical energy storage systems, rechargeable batteries, in other words. Energy storage is critical to stabilizing supply when the power sources are unpredictable, such as solar and wind.

To investigate the issues, the researchers have used a refined version of a computational tool known as NSGA-II, or Non-dominated Sorting Genetic Algorithm II. NSGA-II is a well-established multi-objective optimization algorithm commonly used in engineering and operations research. It is particularly suited to problems with multiple, often conflicting goals, for example, maximizing system reliability while minimizing cost. The novelty in this research is that the team has overcome some of the limitations of standard versions of NSGA-II. In their approach they have extended the algorithm so that it does not settle prematurely on suboptimal solutions, local optima and instead explores all possible outcomes.

This enhancement was achieved first by introducing a new method for selecting the algorithm's starting conditions, its initial population, and secondly by replacing the conventional crossover mechanism with a more exploratory alternative. These enhancements led to an algorithm that was up to five times faster than its predecessor and demonstrably more effective at identifying better solutions across a wider range of scenarios.

The team was able to demonstrate improvements over earlier algorithms in terms of technical performance and economic and investment viability. For instance, one scenario modelled in the study suggested that under specific operational conditions, annual returns on investment in ESPPs could be as high as 13 percent. These findings suggest that, with careful system design and coordination, ESPPs can offer compelling financial incentives in addition to their environmental benefits.

Wu, J., Dong, C., Liu, B., Weng, Z. and Zhu, J. (2025) 'Design of performance evaluation system for electrochemical energy storage power plants based on NSGA-II', Int. J. Power and Energy Conversion, Vol. 16, No. 5, pp.1-20.
DOI: 10.1504/IJPEC.2025.147502

As Europe's ageing population triggers a wave of business ownership transitions, research in the International Journal of Management and Enterprise Development has looked at one of the often-overlooked aspects of business, the factors influencing whether small and medium-sized enterprises (SMEs) can survive such transitions in terms of exit strategies and succession planning.

The researchers studied more than 1200 SMEs and their owner-managers in Finland, all aged 55 or older. They found that planning for the future plays a decisive role in whether firms remain innovative and adaptive as their founders prepare to step down. The research examines how different exit strategies, particularly family succession versus selling to a third party, relate to what ambidexterity. This two-handed ability to exploit existing capabilities while also exploring new opportunities.

Indeed, ambidexterity is widely regarded as essential for a company's long-term viability, particularly in a rapidly changing business landscape. However, achieving it is particularly challenging for SMEs, which often depend heavily on the personal involvement and judgement of a single owner-manager. Without deliberate planning, many small businesses risk stagnation, or even closure, when the boss steps down.

The study's key finding is that firms intending to hand over to a family member are more likely to be ambidextrous, but only if they also engage in strategic planning. In this context, strategic planning refers to a structured process of anticipating and preparing for future developments, including shifts in markets, competition, and internal capacity. Simply having the intention to keep the business in the family does not in itself translate into innovation or preparedness. It is the act of planning, systematically and proactively, that can unlock the firm's ability to balance continuity with change.

In contrast, for businesses with low or no strategic planning, the chosen exit route, whether a family succession or selling to a third party, had little effect on forward-looking or adaptable the business was. This underscores the researchers' argument that the mere expectation of a transition is not enough; it must be accompanied by intentional efforts to chart a path for the firm's future.

Viljamaa, A., Joensuu-Salo, S. and Varamäki, E. (2025) 'Strategic planning, ambidexterity and exit strategy: dynamics in SMEs', Int. J. Management and Enterprise Development, Vol. 24, No. 6, pp.1–18.
DOI: 10.1504/JMED.2025.147533

Operational efficiency often leads to environmental compromise in solid waste management. An approach discussed in the International Journal of Applied Decision Sciences proposes an innovative approach that could improve the collection and transportation of municipal waste. The researchers have turned to the Plant Propagation Algorithm (PPA), a nature-inspired computational method, for a different perspective on routing waste collection vehicles.

Urban sanitation services need vehicles to traverse the city as efficiently as possible. But, working out the schedules and routes many vehicles need to take is no simple planning matter. Vehicles obviously have limited waste capacity, traffic conditions vary, and there are collection time windows with which to contend in terms of access. Moreover, all vehicles need to offload their waste several times on a single shift.

As urban populations grow and waste volumes rise, solving this logistical puzzle becomes increasingly difficult, especially using conventional logistics-planning methods. Many of these traditional approaches are heuristic in nature, which means they are essentially rule-of-thumb. Metaheuristics on the other hand used advanced algorithms based on biological or physical processes. These can efficiently search large sets of possible solutions to find the most viable ones.

The Plant Propagation Algorithm stands out among these tools. It draws its logic from the way strawberry plants reproduce: by sending out runners, some explore nearby soil, others venture further afield, to find ideal growing locations for new plants to grow. This metaphor is converted into a computational search strategy that can balance local and global exploration in the space of potential vehicle routes.

The team created benchmark scenarios using the algorithm and determined performance against criteria such as total distance travelled, fuel usage, the number of vehicles required, and driver working hours. The algorithm was able to generate feasible solutions despite the complexities of urban waste systems anc could handle variable service schedules, diverse waste types, and geographically irregular service zones.

Mat, N.A., Benjamin, A.M., Abdul-Rahman, S., Ku-Mahamud, K.R. and Ramli, M.F. (2025) 'Adaptation of plant propagation algorithm for waste collection vehicle routing problem', Int. J. Applied Decision Sciences, Vol. 18, No. 4, pp.383–407.
DOI: 10.1504/IJADS.2025.147250

Research in the International Journal of Postharvest Technology and Innovation has demonstrated that the knowledge, attitudes, and practices of farmers and traders play a central role in determining how much food is lost after harvest in Kenya's bean supply chains. The insights from the research could have an important effect on food security initiatives across Sub-Saharan Africa.

The research, conducted in Kajiado and Kitui counties in Kenya, examined the supply chains serving the country's Home-Grown School Meals Programme (HGSMP), which provides essential nutrition to schoolchildren using locally sourced food. The focus was on common beans, a crop grown by more than three million small-scale farmers in Kenya and a vital source of protein and nutrients in the region.

While food loss in Sub-Saharan Africa is often discussed in terms of inadequate infrastructure or the impact of pests and diseases, this study turns its attention to human behaviour and specifically knowledge, attitudes, and practices. These include basic yet essential actions such as drying beans properly prior to storage, turning grains during drying, and covering them at night to protect from moisture. Each of these practices can make a measurable difference in preserving food quality and quantity.

The study found that farmers who knew and applied good post-harvest practices saw losses reduced by up to 15%. For traders, proper drying techniques led to nearly 30% less spoilage. Even attitudes, such as the willingness to adopt such drying practices, had a significant effect. The evidence suggests that behavioural change, informed by targeted education and training, could have as much impact as technical interventions such as improved storage facilities.

This shift in focus has implications across Sub-Saharan Africa. As governments and development agencies invest in reducing food loss to address hunger and improve food security, this research shows that equal weight must be given to human behaviour as to infrastructure or technology. In contexts where resources are limited, improving what people know and do may offer a more scalable and cost-effective solution than simply ploughing in more money where it is available.

Okumu, C.A., Mburu, J., Mujuka, E., Ambuko, J. and Klug, I.L.F. (2025) 'Knowledge, attitude and practices relevant to food loss reduction along the bean supply chain of the home-grown school meal program in Kajiado and Kitui counties, Kenya', Int. J. Postharvest Technology and Innovation, Vol. 10, No. 6, pp.1–24.
DOI: 10.1504/IJPTI.2025.147514

Research in the International Journal of Data Mining and Bioinformatics discusses a new approach to demand forecasting for the pharmaceutical retail sector based on an artificial intelligence model. The findings hold promise for improving accuracy in one of the industry's most persistent logistical challenges: managing sales that swing sharply during promotional periods. The new system works better than traditional models by distinguishing between routine demand and the short-term surges driven by marketing campaigns.

The team has built their forecasting system using a machine-learning framework known as the Temporal Fusion Transformer. This deep-learning model is designed specifically to analyse time-series data, such as daily sales figures or seasonal illness rates. Where conventional systems might smooth over the spikes and troughs in this kind of data, the new model can interpret such fluctuations and offer a more nuanced analysis for more reliable forecasting.

One of the underlying factors leading to this improved approach is the model's use of multivariate feature construction. This method can be used to integrate diverse types of data into a single analytical framework. Rather than relying solely on historical sales figures, the model can use public health trends, seasonal disease prevalence, and promotional calendars. By working with such an enriched dataset, the model can detect complex patterns and anticipate demand with much greater precision.

In addition, the system uses a knowledge-guided attention mechanism. This process allows the system to prioritize the most relevant data depending on the sales scenario. For example, during an outbreak of influenza, the model may focus more heavily on regional health reports, whereas during a promotion, it shifts emphasis toward marketing schedules and in-store behaviour. This flexibility allows it to treat routine and promotional demand as fundamentally distinct processes, and so make better predictions about demand.

The researchers have tested their system on data from over 1.2 million retail transactions. The model reduced forecasting errors by almost a quarter compared to traditional methods. When tested in a commercial setting, it led to an almost one-third improvement in medication stock availability and just over a quarter reduction in excess inventory. Such improvements are not merely operational gains. Both outcomes are central to ensuring access to essential medicines while minimising waste in pharmaceutical supply chains.

Zeng, Z., Guo, Y., Ji, Y., Shi, Y. and Feng, T. (2025) 'Data-driven forecasting of pharmaceutical sales: distinguishing promotional vs. daily scenarios', Int. J. Data Mining and Bioinformatics, Vol. 29, No. 5, pp.1–26.
DOI: 10.1504/IJDMB.2025.147534

A new data-driven technique for obstacle avoidance in autonomous vehicles is reported in the International Journal of Vehicle Design. The approach might overcome many of the longstanding challenges in the development of self-driving navigation.

Obstacle avoidance refers to the ability of a vehicle to detect and manoeuvre around objects in its path. Despite years of development, many systems still struggle with this core capability, often producing inefficient routes, reacting slowly to sudden changes, or failing altogether in complex or unpredictable environments. The new method addresses these shortcomings by integrating advanced data mining and optimization algorithms into the vehicle's navigation process.

The researchers explain that multidimensional data mining is key to their approach. This involves extracting patterns from a wide array of data sources, including visual input from cameras, spatial measurements from LIDAR (light detection and ranging, a laser-based system for mapping distances), location data from GPS (Global Positioning System), and real-time traffic information.

This data is processed using K-means clustering, a machine-learning algorithm that groups similar data points without needing prior labels. The purpose is to allow the vehicle to interpret its surroundings more intelligently, recognizing patterns such as obstacle types, road features, or the movement of nearby objects.

Once the environment is processed, the vehicle builds what researchers call a target function. This is a mathematical model that balances the goals of safety, speed, and efficiency. To optimize this function, the team applies the Whale Optimization Algorithm (WOA), a method inspired by the foraging behaviour of killer wales, Orcinus orca. WOA excels at quickly identifying optimal solutions in complex spaces, making it well suited to the high-speed demands of on-road decision-making.

In their simulations, the team demonstrated an obstacle-avoidance success rate of almost 99 percent, with reaction times as fast as 0.44 seconds. These results represent a marked improvement over many existing techniques, which often require longer processing times and produce less direct or more conservative paths.

Wang, A., Yao, Y. and Shang, Z. (2025) 'An obstacle avoidance path selection for autonomous vehicles based on multi-dimensional data mining', Int. J. Vehicle Design, Vol. 97, No. 5, pp.1–21.
DOI: 10.1504/IJVD.2025.147128