MonsoonSIM Enterprise Resource Management Competition 2021

Background:

The MonsoonSIM International ERM Competition (MERMC) is an annual inter-varsity competition hosted by MonsoonSIM and its partners. The competition provides an opportunity for students to gain knowledge of how to adapt to the changing business environments, use the enterprise resource planning (ERP) system, run a successful business with limited resources and manage a large business. It was designed to encourage students to understand the concept of enterprise resource management (ERM) with fun and exciting simulation games, run a virtual enterprise and compete with other virtual companies run by other students.

The MERMC consists of hundreds of fundamental business concepts applicable to trading, distribution, eCommerce, manufacturing, and service business. The concepts are brilliantly wrapped into a typical business simulation with thirteen business departments, including Finance, Retail, Warehouse, Logistics, Planning, Production, Maintenance, Service Management, Human Resources, etc.

Website / Image URL:
https://cretasia.com/zh/%E4%BA%8B%E4%BB%B6%E7%B0%BF/

Award(s):
The Champion – Team name: RADS
1st Runner Up – Team name: Cloudburst
2nd Runner-Up – Team name: Q2fin

Awardee(s):

Team name: RADS

Mr. Ye Jiayuan, BEcon&Fin, Year 2
(3 other team members were students from BASc(FinTech) and BSc, HKU)

 

Team name: Cloudburst

Mr. Chiu Yi Nap, BSc(QFin), Year 1 (photo position: 1st from the left)
Mr. Ng Chun Ki, BEcon&Fin, Year 1 (photo position: 2nd from the left)
(3 other team members were students from the Faculty of Engineering, HKU)

 

Team name: Q2fin

Mr. Chan Yuk Ting, BSc(QFin), Year 2
Miss Lam Yu Yan, BSc(QFin), Year 2
Miss Shum Wing Yan , BSc(QFin), Year 2
Mr. Wang Jinbing , BSc(QFin), Year 2
Miss Yiu Sing Yi, BSc(QFin), Year 2

 

Students Sharing:
“There are basically two things I learnt from the competition.

First we were having a bad time in some of the practice games, we had a very unsatisfied ranking. And our team had serious disagreements on some strategies. We even argued our ideas until 2 or 3 o’clock at midnight. Most importantly, the ultimate goal of our argument was our performance in the competition, all of us wanted to make ourselves better. After detailed communication and subsequent adjustments and tests, we finally found the strategy that worked best for us. And we got a satisfactory result in the end. Therefore, don’t mind the quarrel, as long as the members are for good performance, effective and calm communication will bring good results.

Second, to be honest, we never expected to win the championship. We thought we have the strength and potential to win the championship but there was a team which was killing the game. We had never seen them not take first place in all the practice games. Compared to them, we can only say that we had the strength of the top three, but our ranking was not stable in so many exercises. But we were always learning, always adjusting, always experimenting with new directions and practicing details. This allowed us to have our best status in the final and win the championship. Therefore, do not be afraid of the opponent’s strength, learn from them, and insist on strengthening yourself, nothing is impossible to cross.”

(by Ye Jiayuan)

“MonsoonSIM Enterprise Resource Management Competition (MERMC) is an annual inter-varsity event where teams simulated to run a virtual company and competed for the highest business KPI targets set on the spot. Each team needed to cooperate and handle the routine operation of nearly ten departments throughout the simulation. After 2 months of training, teams needed to compete against one another for the highest business KPI. Additionally, in the finals, teams would need to present and elaborate on their business model as well as how their tactics stay in line with ESG principles.
Not only had our team received hands-on experience in applying theories in our actual operation, but we had also become much more adaptable and cooperative with one another thanks to the format of this competition. MERMC provided an opportunity for us to explore our limits with an unyielding determination to improve.”
(by Chiu Yi Nap and Ng Chun Ki)

“Throughout the competition, we first learned the importance of cooperation. Everyone had different processes to monitor, and these were usually highly interdependent. Therefore, we had to communicate frequently with each other to ensure a comprehensive business strategy with an efficient resource allocation.
Secondly, since there were some unexpected events like financial crises happened in the middle phase of the game, we had to make quick decisions ranging from product procurement, warehouse management, and production, which taught us to be more flexible in making company-wide decisions.
Lastly, this game taught us about the trade-off of a company. The scoring matrix of the game consists of some conflicting indicators such as production and pollution, while it was impossible to achieve high scores in both categories. So, the only solution would be to give up some scores in one aspect for better performance of the others to achieve benefit maximization.
Lastly, as the theme for this year’s competition is ESG, we also learned about how important the sustainable development of a business is and why companies need to fulfill them for the well-being of the whole society.”
(by Chan Yuk Ting, Wang Jinbing, Shum Wing Yan, Lam Yu Yan and Yiu Sing Yi)

Other Events
港大经管学院最新多模态AI图像生成能力排名出炉 部份中国人工智能模型表现突出
2025 | 学院新闻
港大经管学院最新多模态AI图像生成能力排名出炉 部份中国人工智能模型表现突出
港大经管学院今日发表《人工智能模型图像生成能力综合评测报告》,针对15个“文生图模型”及7个“多模态大语言模型”进行全面评估。研究显示,字节跳动的即梦AI和豆包,以及百度的文心一言,在新图像生成的内容质素及图像修改的表现突出;而早前引起全球关注的DeepSeek最新推出的文生图模型Janus-Pro,则在新图像生成方面表现欠佳。研究亦发现部分文生图模型虽然在内容质素方面表现优异,却在安全与责任方面的表现强差人意。整体而言,与文生图模型相比,多模态大语言模型整体表现较佳。
人工智能图像生成评测成绩单公布: 字节跳动百度表现亮眼,DeepSeek Janus-Pro表现欠佳
2025 | 研究
人工智能图像生成评测成绩单公布: 字节跳动百度表现亮眼,DeepSeek Janus-Pro表现欠佳
如今,人工智能领域的前沿模型技术已经从文本处理拓展至视觉信息的深度理解与生成。这些模型既能精准解读图像语义,又能根据文字描述创作出兼具真实感与艺术性的视觉内容,展现出令人惊叹的跨模态理解与创作能力。本研究聚焦全新图像的生成和基于现有图像的图像修改两大核心任务,提出了一套系统性的人工智能模型图像生成能力评测框架。我们基于多维测试集的构建与专家评审,对15个专业文生图模型和7个多模态大语言模型的图像生成能力进行了全面评估。结果显示,字节跳动的即梦AI和豆包以及百度的文心一言在新图像生成的内容质量与修改任务中表现突出,位列第一梯队。对比不同类型的AI模型,我们发现,相对于专业文生图模型,多模态大语言模型整体表现更佳。