Connect with us
We’re experimenting with AI-generated content to help deliver information faster and more efficiently.
While we try to keep things accurate, this content is part of an ongoing experiment and may not always be reliable.
Please double-check important details — we’re not responsible for how the information is used.

Civil Engineering

Faster Planning for Complex Problems with Machine Learning

Researchers developed a machine-learning-guided technique to solve complex, long-horizon planning problems more efficiently than some traditional approaches, while arriving at an optimal solution that better meets a user’s goals.

Avatar photo

Published

on

The faster planning system developed by MIT researchers uses machine learning to reduce solve time by up to 50 percent and produce a solution that better meets a user’s objective, such as on-time train departures. The new method could also be applied to other complex logistical problems like scheduling hospital staff or assigning airline crews.

Engineers often break down these kinds of problems into a sequence of overlapping subproblems that can each be solved in a feasible amount of time. However, the overlaps cause many decisions to be needlessly recomputed, making it take much longer to reach an optimal solution.

The researchers’ new approach learns which parts of each subproblem should remain unchanged and freeze those variables to avoid redundant computations. A traditional algorithmic solver then tackles the remaining variables.

“This is a very complex combinatorial scheduling problem,” says Cathy Wu, a member of the Laboratory for Information and Decision Systems at MIT. “Our approach can be applied without modification to all these different variants.”

The researchers’ technique, which they call learning-guided rolling horizon optimization (L-RHO), teaches a machine-learning model to predict which operations should be recomputed when the planning horizon rolls forward.

To test their approach, the researchers compared L-RHO to several base algorithmic solvers and specialized solvers. It outperformed them all, reducing solve time by 54 percent and improving solution quality by up to 21 percent.

Their method continued to outperform all baselines even when tested on more complex variants of the problem, such as factory machines breaking down or extra train congestion.

“Our approach can be applied without modification to all these different variants,” says Wu. “It even outperformed additional baselines we created to challenge our solver.”

L-RHO can also adapt if the objectives change, automatically generating a new algorithm to solve the problem – all it needs is a new training dataset.

In the future, the researchers want to better understand the logic behind their model’s decision to freeze some variables, but not others. They also want to integrate their approach into other types of complex optimization problems like inventory management or vehicle routing.

This work was supported by the National Science Foundation and MIT’s Research Support Committee, among others.

Biochemistry

A Game-Changing mRNA Vaccine that’s More Effective and Less Costly to Develop

A new type of mRNA vaccine is more scalable and adaptable to continuously evolving viruses such as SARS-CoV-2 and H5N1, according to a new study.

Avatar photo

Published

on

A team of researchers from the University of Pittsburgh School of Public Health and Pennsylvania State University has made a groundbreaking discovery in the field of vaccine development. They have created a new type of mRNA vaccine that is not only more effective but also less costly to develop, making it a game-changer in the fight against infectious diseases.

The current mRNA vaccines, such as those used to prevent COVID-19, have two significant challenges: they require a high amount of mRNA to produce and are constantly evolving due to the changing nature of viruses like SARS-CoV-2 and H5N1. The researchers addressed these challenges by creating a proof-of-concept COVID-19 vaccine using what’s known as a “trans-amplifying” mRNA platform.

In this approach, the mRNA is separated into two fragments: the antigen sequence and the replicase sequence. The latter can be produced in advance, saving crucial time in the event of a new vaccine needing to be developed urgently and produced at scale. Additionally, the researchers analyzed the spike-protein sequences of all known variants of SARS-CoV-2 for commonalities, rendering what’s known as a “consensus spike protein” as the basis for the vaccine’s antigen.

The results are promising: in mice, the vaccine induced a robust immune response against many strains of SARS-CoV-2. This has the potential for more lasting immunity that would not require updating, because the vaccine has the potential to provide broad protection. Additionally, this format requires an mRNA dose 40 times less than conventional vaccines, so this new approach significantly reduces the overall cost of the vaccine.

The lessons learned from this study could inform more efficient vaccine development for other constantly evolving RNA viruses with pandemic potential, such as bird flu. The researchers hope to apply the principles of this lower-cost, broad-protection antigen design to pressing challenges like bird flu, making it a crucial step in preparing for future pandemics.

Continue Reading

Batteries

Unlocking the Potential of Solid-State Batteries

Researchers have discovered that the mixing of small particles between two solid electrolytes can generate an effect called a ‘space charge layer,’ an accumulation of electric charge at the interface between the two materials. The finding could aid the development of batteries with solid electrolytes, called solid-state batteries, for applications including mobile devices and electric vehicles.

Avatar photo

Published

on

By

The development of solid-state batteries has been gaining momentum in recent years, promising safer and more powerful alternatives to traditional lithium-ion batteries. A team of researchers from the University of Texas at Dallas has made a significant breakthrough in this field by discovering that mixing small particles between two solid electrolytes can generate an effect called a “space charge layer.” This accumulation of electric charge at the interface between the materials has been found to create pathways that make it easier for ions to move across, potentially leading to better-performing solid-state batteries.

The researchers, led by Dr. Laisuo Su and Dr. Kyeongjae Cho, published their study in ACS Energy Letters, where it was featured on the cover of the March issue. They discovered that when the separate solid electrolyte materials make physical contact, a layer forms at their boundary where charged particles, or ions, accumulate due to differences in each material’s chemical potential.

“Imagine mixing two ingredients in a recipe and unexpectedly getting a result that is better than either ingredient alone,” Dr. Su explained. “This effect boosted the movement of ions beyond what either material could achieve by itself.”

The research is part of the university’s Batteries and Energy to Advance Commercialization and National Security (BEACONS) initiative, which aims to develop and commercialize new battery technology and manufacturing processes. The team’s findings suggest a new way to design better solid electrolytes by carefully choosing materials that interact in a way that enhances ionic movement.

Solid-state batteries show promise for generating and storing more than twice as much power as batteries with liquid electrolytes, while being safer because they are not flammable. However, the development of solid-state batteries faces challenges due to difficulties in moving ions through solid materials.

The researchers plan to continue studying how the composition and structure of the interface lead to greater ionic conductivity. This breakthrough has the potential to unlock the full potential of solid-state batteries, enabling advanced battery systems that can improve the performance of drones for defense applications.

In conclusion, the discovery of the space charge layer phenomenon offers a promising new direction for the development of solid-state batteries. By understanding and harnessing this effect, researchers may be able to create more efficient and powerful batteries that meet the growing demands of mobile devices, electric vehicles, and other applications.

Continue Reading

Ancient Civilizations

Reviving an Ancient Hue: Researchers Recreate Egyptian Blue Pigment

Researchers have recreated the world’s oldest synthetic pigment, called Egyptian blue, which was used in ancient Egypt about 5,000 years ago.

Avatar photo

Published

on

By

The world’s oldest synthetic pigment, Egyptian blue, has been recreated by a team of researchers from Washington State University. This breakthrough, published in the journal NPJ Heritage Science, provides valuable insights for archaeologists and conservation scientists studying ancient Egyptian materials.

Led by John McCloy, director of WSU’s School of Mechanical and Materials Engineering, the research team collaborated with the Carnegie Museum of Natural History and the Smithsonian’s Museum Conservation Institute to develop 12 recipes for the pigment. These recipes utilized a variety of raw materials and heating times, replicating temperatures available to ancient artists.

Egyptian blue was highly valued in ancient times due to its unique properties and versatility. It was used as a substitute for expensive minerals like turquoise or lapis lazuli and applied to wood, stone, and cartonnage – a papier-mâché-type material. Depending on its ingredients and processing time, the pigment’s color ranged from deep blue to dull gray or green.

The researchers’ work aimed to highlight how modern science can reveal hidden stories in ancient Egyptian objects. After the Egyptians, the pigment was used by Romans, but by the Renaissance period, the knowledge of how it was made had largely been forgotten.

In recent years, there has been a resurgence of interest in Egyptian blue due to its intriguing properties and potential new technological applications. The pigment emits light in the near-infrared part of the electromagnetic spectrum, which people can’t see, making it suitable for fingerprinting and counterfeit-proof inks. It also shares similar chemistry with high-temperature superconductors.

To understand the makeup of Egyptian blue, the researchers created 12 different recipes using mixtures of silicon dioxide, copper, calcium, and sodium carbonate. They heated the material at around 1000 degrees Celsius for between one and 11 hours to replicate temperatures available to ancient artists. After cooling the samples at various rates, they studied the pigments using modern microscopy and analysis techniques that had never been used for this type of research.

The researchers found that Egyptian blue is highly heterogeneous, with different people making the pigment and transporting it to final uses elsewhere. Small differences in the process resulted in very different outcomes. In fact, to get the bluest color required only about 50% of the blue-colored components, regardless of the rest of the mixture’s composition.

The samples created are currently on display at Carnegie Museum of Natural History in Pittsburgh, Pennsylvania and will become part of the museum’s new long-term gallery focused on ancient Egypt. This research serves as a prime example of how science can shed light on our human past, revealing hidden stories in ancient objects and materials.

Continue Reading

Trending