Multi-Objective Optimization of Gating System Design in Castings for Yield and Sustainability Improvement

In the world of manufacturing, few processes are as ancient or as vital as casting. It is the method by which molten metal is poured into a hollow mold, allowed to cool, and then revealed as a solid object, ranging from tiny engine parts to massive ship components. For centuries, foundries have relied on the skill of experienced workers to design the channels that guide the hot metal into the mold and the reservoirs that feed it as it shrinks while cooling. These channels, known as the gating system, are critical. If they are too small or poorly placed, the metal flows turbulently, trapping air or creating weak spots. If the feeding reservoirs are too large, they waste valuable metal; if too small, the final part may crack or crumble as it solidifies. The goal has always been a delicate balance: to produce a perfect, defect-free part while using as little extra metal as possible, because every drop of wasted metal must be melted down again, consuming energy and resources.

A team of researchers from India has tackled this enduring challenge by applying modern computer simulation to the design of a specific industrial part: a grey iron rotor used in heavy machinery. Rather than relying solely on trial and error, which involves pouring metal, cutting parts open to check for flaws, and starting over, the team built a detailed digital model of the casting process. They focused on a material called FG260 grey cast iron, a strong and durable metal commonly used for brake rotors and structural components. The researchers created three different designs for the channels and feeding reservoirs that would surround the rotor. Using powerful software, they simulated how the molten metal would flow into the mold, how it would cool, and where it might shrink or form holes as it turned from liquid to solid. This allowed them to predict the outcome of each design before a single drop of metal was poured in a real factory.

The study revealed a clear trade-off that has long plagued foundry engineers. One of the designs they tested, which used the smallest amount of extra metal, produced the highest yield—the ratio of the useful part to the total metal poured. However, the simulation showed that this design failed to feed the thick center of the rotor properly, leaving it with internal shrinkage defects that would weaken the part. Another design used large feeding reservoirs to ensure the metal flowed perfectly and the part remained solid, but this approach wasted a significant amount of material, lowering the yield. The third design, which the researchers ultimately selected, struck a precise balance. It used just enough feeding metal to eliminate all internal defects while keeping the waste to a minimum. In the computer simulation, this optimized layout achieved a yield of 88.39 percent, meaning that nearly 89 percent of the metal poured ended up in the final product, with the rest being the necessary channels and reservoirs. This was a significant improvement over the factory's current method, which yielded only 82.78 percent.

To ensure these digital predictions were not just theoretical, the researchers took their best design to a real foundry. They poured actual metal into molds built with the new gating system and compared the results to the computer models. The physical tests confirmed the simulation's accuracy. The real castings showed the same pattern of solidification and the same lack of defects as the virtual ones. The experimental yield came in at 87.33 percent, a figure very close to the simulated 88.39 percent, proving that the computer model could reliably guide real-world production. The difference between the old method and the new one might seem small in percentage points, but the impact on a large scale is substantial. For a factory producing 50,000 of these rotors a year, the improved design saves 32.5 metric tons of metal. This reduction in material use translates directly into energy savings, as less metal needs to be melted, and a reduction in carbon dioxide emissions by nearly 14.6 metric tons annually.

The financial benefits are equally tangible. By saving 32,500 kilograms of metal each year, the factory avoids the cost of purchasing and melting that material, resulting in a direct annual saving of approximately 24.37 lakh rupees. Beyond the money, the process becomes cleaner and more efficient, with less waste to manage and less strain on the furnaces. The researchers demonstrated that by treating the design of these channels as a multi-objective problem—where both the quality of the part and the efficiency of the material use must be optimized simultaneously—foundries can move beyond guesswork. They showed that a systematic approach, combining detailed computer analysis with real-world testing, can identify a configuration that is superior to traditional methods. This work does not just offer a better way to make one specific rotor; it provides a reproducible framework that other foundries can use to improve their own processes, turning a routine manufacturing step into a precise science that saves money, energy, and the environment.