Integrating CRITIC and Weighted K-Means for Humanitarian Relief Center Location Planning: Evidence from Gaziantep
This study proposes a hybrid decision-support framework that integrates the CRITIC method with a weighted k-means algorithm to optimize the strategic placement of humanitarian relief centers in park areas within Gaziantep, ensuring needs-based and objective planning for post-disaster scenarios.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
When a disaster strikes, the race to save lives often hinges on a single, deceptively simple question: where do we put the aid? Humanitarian relief is not just about having enough food, water, and medicine; it is about getting those supplies to the people who need them before time runs out. This challenge belongs to a field known as humanitarian logistics, which treats the movement of aid as a complex puzzle of geography, risk, and human need. If a distribution center is placed too far from a crowded neighborhood, people wait too long. If it is placed too close to a gas station or a fault line, the center itself could become a target for fire or collapse. The goal is to find a spot that balances the urgent demand of the population with the safety of the location, a task that becomes incredibly difficult when the ground beneath a city is unstable and the data is incomplete.
In a recent study focused on the city of Gaziantep in Turkey, researchers tackled this problem by combining two distinct ways of thinking: one that weighs the importance of different factors, and another that groups similar areas together to find patterns. The team, led by Zeynep Yüksel, İbrahim Miraç Eligüzel, and Süleyman Mete, looked at the city's park areas as potential sites for temporary relief centers. Parks are logical candidates because they offer open space that is less likely to be buried under rubble from collapsed buildings. However, not all parks are equal. Some are near dense housing where the need for aid is highest, while others might be dangerously close to earthquake faults or fuel depots. To solve this, the researchers developed a two-step process. First, they used a method called CRITIC to objectively decide which factors mattered most, letting the data itself determine the weight of each criterion rather than relying on human opinion. Second, they fed those weights into a machine learning tool called weighted k-means, which acted like a digital organizer, sorting the 57 candidate parks into groups and identifying the best central spot for each group to serve as a relief hub.
The researchers examined six specific factors to judge each park. They looked at how many people lived nearby, how many buildings surrounded the park, the size of the park itself, and how close it was to hospitals, schools, and other social facilities. Crucially, they also treated three factors as risks: how close the park was to a fuel station, how close it was to a known earthquake fault line, and how dense the surrounding buildings were, as heavy debris from tall structures could block access. By analyzing real data from Gaziantep's Şahinbey district, the team found that safety was the most critical driver. The data showed that proximity to fuel stations and proximity to fault lines carried the highest importance, meaning that avoiding danger was more significant than simply being close to a large crowd. This objective finding challenged the assumption that the most populated area is always the best place for a center; instead, the study suggested that a slightly less crowded but safer location is often the smarter choice.
Once the importance of these factors was established, the team tested different scenarios to see how many relief centers the city would need. They simulated situations with three, four, five, and six centers to see how the distribution of aid would change. When they modeled a system with only three centers, the results were uneven. One center was forced to serve a massive area with 31 different demand points, creating a bottleneck that would likely slow down the delivery of supplies. As the team increased the number of centers to four, the load became much more balanced. The service areas shrank, the distances people had to travel decreased, and the overall efficiency of the network improved significantly. The study found that moving from three to four centers reduced the total distance within the system by more than 40 percent, a massive gain in a situation where every minute counts.
However, the researchers also warned that adding more centers is not always the answer. When they tested scenarios with five and six centers, the network became fragmented. While the distances to individual points became even shorter, some centers ended up serving only two or three locations, which would be an inefficient use of limited resources like vehicles and staff. The statistical analysis confirmed that the difference between these scenarios was real and significant, proving that the number of centers chosen changes the entire character of the relief operation. The study concluded that there is no single perfect number of centers. Instead, decision-makers need a flexible framework that allows them to choose based on their available resources. If supplies are scarce, a smaller network with fewer, larger centers might be necessary. If the priority is speed and access, a larger network with more centers is better, provided there are enough resources to manage them.
This research offers a clear path forward for city planners and emergency managers. By using a method that lets the data speak for itself, rather than guessing which factors are important, planners can avoid the biases that often lead to poor decisions in the chaos of a disaster. The study demonstrates that the best location for a relief center is not just a matter of geography, but a calculation of risk, demand, and capacity. In the aftermath of a catastrophe, where the ground may still be shaking and the roads may be blocked, having a plan that balances safety with speed could mean the difference between a community that recovers quickly and one that suffers needlessly. The work in Gaziantep provides a blueprint for how to make those critical choices with clarity and confidence, ensuring that when help arrives, it arrives where it is needed most, and where it can do the most good.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.