Urban transportation significantly contributes to global carbon emissions, thus emphasizing the importance of green and efficient alternatives like public transportation. Enhancing the appeal of public transport through the strategic placement of bus stops can mitigate carbon emissions concurrently. This research proposes a Quantum Entangled Self-Attention Neural Networks (QESANN) model that considers both the passengers and stops characteristics to optimize bus stop locations. The QESANN model integrates the complexity of the relationship between passengers and bus stops characteristics, thereby fostering a more holistic approach to bus stop placement. It uses a Conditional Random Field (CRF) to initialize and process the attributes of both commuters and bus stops through quantum entanglement. The Vehicle Specific Power (VSP) model is employed to gage the variation in bus carbon emissions pre and post bus stop location optimization. Simulation tests, conducted via the Anylogic software and the IBM Quantum platform, demonstrated promising results. Over a yearly travel cycle for 1,000 individuals, the number of bus commuters rose from 322 to 651, concurrently reducing urban traffic carbon emissions by 430,243.70 kg annually. These findings affirm the effectiveness of the proposed QESANN model in enhancing the appeal of bus stops, reducing carbon emissions from urban transportation, and fostering a more sustainable urban environment.

中文翻译：

---

量子纠缠和自注意力神经网络：对乘客和车站特征的研究以实现最佳公交车站定位


城市交通对全球碳排放做出了重大贡献，因此强调了公共交通等绿色高效替代方案的重要性。通过战略性地布局公交车站来增强公共交通的吸引力可以同时减少碳排放。这项研究提出了一种量子纠缠自注意力神经网络（QESANN）模型，该模型考虑乘客和车站特征来优化公交车站位置。 QESANN 模型整合了乘客与公交车站特征之间关系的复杂性，从而促进了更全面的公交车站布局方法。它使用条件随机场（CRF）通过量子纠缠来初始化和处理通勤者和公交车站的属性。车辆特定功率 (VSP) 模型用于衡量公交车站位置优化前后公交车碳排放的变化。通过 Anylogic 软件和 IBM Quantum 平台进行的模拟测试显示出有希望的结果。在1000人每年出行周期中，公交车通勤人数从322人增加到651人，同时每年减少城市交通碳排放430,243.70公斤。这些发现证实了所提出的 QESANN 模型在增强公交车站吸引力、减少城市交通碳排放以及培育更加可持续的城市环境方面的有效性。