On December 5th, 2020, the final of the "Alibaba Cloud Smart Service Innovation Challenge - Service Scheduling Competition" hosted by Alibaba and Zhejiang University successfully came to an end in Hangzhou. The team from SIAR was composed of DING Juntao, LIU Weihong, ZHAI Wenjie and TANG Xin. These four students enrolled in 2019, who were mentored by ZHU Zongwei from the Intelligent Cloud Computing Center, Suzhou Institute of Advanced Research (SIAR), won the bronze medal in the competition.

Fig. 1 The Award Ceremony

The Service Scheduling Competition was valued by the industry and academia with the presence of LI Jin, Vice President of Alibaba Group, Prof. YANG Genke, Executive Director of Ningbo Institute of Artificial Intelligence, Shanghai Jiao Tong University, Prof. SHAO Zhijiang, Dean of College of Control Science and Engineering of Zhejiang University, Director of Global Technical Service Department of Alibaba Cloud Intelligence, Senior Algorithm Expert at Machine Intelligence Technology of Alibaba DAMO Academy, Senior Algorithm Expert of Artificial Intelligence Department, Cainiao Smart Logistics Network, and a number of professors from Zhejiang University as judges of the competition.

Fig.2 The Team from SIAR, USTC

In the setting of real application scenario, the competition exhibited the full-cycle technical services provided by Alibaba Cloud for many enterprises, government agencies and developers around the world. Among all these services, service scheduling optimization is one of the core elements to improve the customer experience. However, in the face of huge service volume, vastly different regions and languages, various customer problems, complex service scenarios, and technicians with different strengths and weaknesses, it is a tough task to build the service scheduling optimization algorithm for the rapid distribution of massive services while efficiently matching service scenarios with technical experts' strengths. The service scheduling competition focuses on the integration of industry, research and academia, which provides valuable experience for the development of China's "digital intelligence service" industry.

The competition lasted three months and attracted 1,381 teams from over 100 universities such as Tsinghua University, Peking University, University of Science and Technology of China, Zhejiang University, Shanghai Jiao Tong University, Nanjing University, Xi'an Jiaotong University, Huazhong University of Science and Technology, National University of Defense Technology, and other major enterprises to participate. The team from SIAR adopted a process scheduling approach innovatively combining with particle swarm optimization (PSO) and shallow neural network, constructed a new operator to replace the complicated nonlinear relationship of deep network, and used PSO to overcome the difficulty of Back Propagation (BP) optimization in NP-complete. The team entered the finals with the sixth place in the preliminary round and the third place in the final round, and finally won the bronze medal with the fourth place.

Fig. 3 The Group Photo of Organizers and Participants