October 1998 China: Ancient Country Makes OR History China is in the process of transferring from a planned economy to a market-driven one, creating opportunities and challenges for the massive nation's active operations research community. A look at the state of OR in China as Beijing prepares to host next year's IFORS 15th Triennial Conference By Xiang-Sun Zhang and Kan Cheng Modern operations research activities in China were initiated in the late 1950s. The first OR group was founded at the Institute of Mechanics, Chinese Academy of Sciences (CAS), in 1956, promoted by professors H.S. Tsien and K.C. Hsu. Tsien received his master's degree from MIT and his Ph.D. from the California Institute of Technology, and went on to become Cal Tech's first Goddard Professor. Hsu earned his Ph.D. at the University of Kansas, and was a research associate in the Institute of Fluid Mechanics and Applied Mathematics at the University of Maryland. Tsien and Hsu returned to China in 1955. In 1959, a second OR division was set up at the Institute of Mathematics, CAS. These two divisions merged into one division in 1960 as a part of the Institute of Mathematics. The main research interests at that time were queueing theory, nonlinear programming and graph theory along with transportation theory, dynamic programming, quality control and economic analysis. In 1963, the division organized a series of specialized courses in OR at the Chinese University of Science and Technology, marking the first time in China that a university offered a systematic education in OR. Today, OR courses are a basic requirement in the business schools and engineering departments of almost every university in China. Early OR application activities (late 1950s) in China focused on transportation problems. One such example was the allocation design of threshing grounds to save time and manpower. The classic Chinese Postman Problem model was presented by Professor M.G. Guan in the same period. One of the early highlights of OR application activities in China was conducted by the Chinese mathematician L.G. Hua — a member of CAS and president of the Chinese Society of Mathematics — during the so-called Great Cultural Revolution. Since theoretical research had been stopped at that time, Hua formed his own small group and turned to teaching basic optimization techniques at countryside factories in an effort to help managers and engineers apply OR in their daily operations. For a decade beginning in 1965, Hua visited more than 20 provinces and innumerable towns and cities with his group and planted the seeds of OR principles and philosophies wherever he went. His work promoted the development of OR in China then and continues to this day, long after his death. The Operations Research Society of China (ORSC) was founded after the Great Cultural Revolution in 1980 when the First National Conference of OR was held in Shandong Province (on the east coast). Hua was elected the first president of ORSC. ORSC became a member of IFORS in 1982. Professor M.Y. Yue assumed the presidency of ORSC at the Second National Conference held in 1984 in Shanghai, the commercial center of China. The Third National Conference was given in Anhui Province in the middle of the mainland in 1988. The Fourth National Conference was hosted by several universities in Sichuan Province in the southwest area of China near the upper reaches of the Yangtse River. Professor G.H. Hsu served as president of ORSC from 1988-96. The Fifth National Conference, held at the ancient city of Xian in 1996, marked the beginning of Professor X.S Zhang's leadership of ORSC. For the past decade, ORSC has been active in the development of the Association of Asian-Pacific Operational Research Societies (APORS) within IFORS. As the president of APORS from 1991 to 1994, Hsu organized the Second Conference of APORS in Beijing in the fall of 1991. As a representative of APORS, Hsu also served as vice president of IFORS from 1992-94. Organization of ORSC The ORSC now has 500 qualified members and a dozen chapters located throughout the vast country. The most recently established chapter, the Daqing branch, is noteworthy for several reasons. First, Daqing is the largest oil industrial base in China and home to 500 of the biggest companies in the nation. Second, the Daqing branch is a milestone of OR application in China since most of its members come from industry, unlike the majority of the ORSC members who hail from universities and research institutes. Academically, there are several special committees under ORSC. The list includes: Mathematical Programming. Queuing Theory and Applications. Reliability Theory. Decision-Making Theory and Applications. Simulation and Information Techniques. Intelligence Computing. Scheduling Theory. OR Education. The special committees organize symposiums every year or every other year during the interval of the two National Conferences. Many of the committees and local branches present prizes recognizing outstanding work. To promote the cooperation with OR colleagues in the Asia-Pacific area and the rest of the world, the Asia-Pacific Operations Research Center (APORC) was set up in 1995 in Beijing. The center is affiliated with the CAS and APORS. APORC has already organized three symposiums entitled "International Symposium on Operations Research and Applications in Engineering, Technology and Management (ISORA)." The first was held in 1995 in Beijing; the second was held in 1996 in Kweilin, one of the most picturesque cities in the world; the third was held in August in Kunming. The symposium Proceedings are published under the title "Lecture Notes in Operations Research." Status of Academic Research and Application of OR in China During the 1980s many universities and colleges started to set up OR courses in their engineering schools, business schools or economic management schools. Linear programming, nonlinear programming, networks, queueing theory, reliability theory, inventory and other major OR techniques are all taught at various levels. OR degrees (M.S. and Ph.D.) are offered in some universities and institutes such as Tsinghua University, Chinese University of Science and Technology, Fudan University, Institute of Applied Mathematics within CAS and the Institute of System Sciences within CAS. ORSC has two official journals: OR Transactions, a quarterly journal that publishes original articles in OR and related fields (the language can be in Chinese or English); and OR and MS (in Chinese only), a quarterly aimed at OR practitioners. Two journals sponsored by CAS — Acta Mathematicae Applicatae Sinica and System Science and Mathematics — also include OR-related papers. There are two major avenues for OR researchers in China to obtain grants. The first is through the National Natural Science Foundation (NNSF), which has given about one-third of its total funds for mathematicians to researchers involved in OR mathematics. Researchers can also apply for grants from other branches of the NNSF such as Management, Information Science and Automation. The second means to receive grant money is by providing consulting services to the public sectors such as the central government, local government, industrial sectors and so on. In these applications, traditional OR models and algorithms are built into software, and sometimes a complete management information system or a decision support system with OR-type philosophy and algorithms at its core part are provided to the user. One such example, a "Project Evaluation System in the State Economic Information System of China," by Professor X.S. Zhang and J.C. Cui from the OR Division of Institute of Applied Mathematics, CAS, won the IFORS Prize for OR in Developing Countries at the 14th Triennial Conference in Vancouver, Canada. The core part of the project is a reverse model of Data Envelopment Analysis (DEA). Most OR activities in China have their foothold in industrial and agricultural applications, such as applications in petrochemical enterprises, manufacture engineering and structure design of products. Some OR practitioners have focused their attention on supporting high-level government decision-makers in their strategic planning about the country's continued advance in economy and ecology. The project, "Optimization of the Ecological Economic Development for the Upper Reaches of the Yangtse River," by professor G.Z. Liu and others from Chengdu University of Science and Technology, was awarded the second place prize in the IFORS Prize for OR in Developing Countries competition. Another OR research project which had great social influence in China is "Grain Output Forecasting" conducted by Professor X.K Chen from the Institute of System Science, CAS. With an input-occupancy-output technique at its core, the project has greatly increased the accuracy of the national grain output forecast. The group and its achievement received accolades from top government officials as well as CAS, which awarded the group its highest prize. IFORS'99 in Beijing China is now in the process of transferring from a planned economy to a market-oriented economy. It offers opportunities and challenges for China's OR community. It is ORSC's pleasure to host the IFORS 15th Triennial Conference in Beijing, since it is an opportunity for us to learn the new developments and trends of OR in other countries, especially in the developed countries. With the help of the IFORS executives, the preparation for next year's IFORS conference is proceeding smoothly. We are looking forward to welcoming the world's OR community to this great event. Beijing to Host IFORS '99 IFORS '99, the 15th Triennial Conference of the International Federation of Operational Research Societies, will be held Aug. 16-20, 1999, in Beijing, China. Hosted by the Operations Research Society of China, the conference will be built around the theme, "OR — Parallel Roads to Prosperity in the 21st Century." Conference topics will range from global manufacturing, marketing, networking and optimization to reliability, risk analysis and routing. More than 30 session clusters will be organized by prominent researchers and practitioners. In addition, more than 50 invited sessions will be prepared along with several tutorials on specific topics. Graham Rand is organizing a plenary session marking the 40th anniversary of IFORS, and national contributions are expected from various society members of IFORS. Individuals interested in organizing sessions (with three or four contributions each) are invited to contact Program Committee Chairman Paolo Toth of D.E.I.S.-University of Bologna (E-mail: ifors99_pc@deis.unibo.it). Nov. 30, 1998 is the deadline for submission of abstracts by mail or fax; Dec. 31 is the deadline for electronic submission of abstracts. Optional submission of full papers to ITOR are due May 31, 1999. One of the conference's many highlights will be the awarding of the "OR in Developing Countries Prize." The conference will be held in the historical city of Beijing which is quickly developing into a modern metropolis. The Great Wall, the Forbidden City, the Temple of Heaven and the Summer Palace are just a few of the sights that await visitors. Special tours will be organized before and after the meeting for those interested in visiting other cities, and day tours will be available during the meeting for spouses. A travel agency will take care of the reservations. The Friendship Hotel, one of the largest garden-style hotels in Asia, will serve as the conference headquarters. Professor Xiang-Sun Zhang of the Institute of Applied Mathematics-Chinese Academy of Sciences is chairman of the Organizing Committee, while Professor Kan Cheng of the Institute of Applied Mathematics-Chinese Academy of Sciences serves as the Conference Secretariat. The IFORS '99 Conference URL is http://www.ifors.org/conferences/ifors99/; the Organizing Committee's e-mail address is orschina@public.east.cn.net. Case Study: Prediction for China Grain Output By Xikang Chen and Xiaoming Pan Feeding more than 1.2 billion people is — and for the foreseeable future will be — the first priority of China. More than 330 million rural Chinese laborers are involved in the production of grain. Needless to say, it is of great importance to the country and its people to accurately forecast grain output so as to best arrange grain consumption, storage, import and export. Up until 1980 China used three main approaches for predicting grain yield and output: 1. meteoric, 2. remote-sensing, and 3. biological statistical simulation. These approaches generally produced forecasts with 5 to 10 percent error rates while looking no more than two months forward in time. Looking at the approaches individually, the error rate of the weather forecast beyond a half a year was more than 50 percent; the forward time of remote sensing to assess crop lands and yields was less than one month; and the comprehensive biological model concerning sunshine, temperature, precipitation and other factors to predict crop yield could only be used on small, experimental land areas because it was impossible to accurately get all data needed across a vast region like China. The government needed accurate forecasts looking at least six months forward to readily deal with the grain issues. In order to alleviate the problem, a research group of the Institute of Systems Science (ISS), under the Chinese Academy of Sciences (CAS) and headed by Professor Xikang Chen, developed a systematic integrated approach, formally known as the "Social, Economic and Technological Approach for China Grain Output." The approach takes into account the influences of agricultural inputs, technological advancements, natural factors such as climate and pests, and policies on grain output. The key techniques of this approach are: 1. input-occupancy-output analysis (an extension of classical input-output analysis), 2. variable coefficient forecasting equation with the consideration of the diminishing return law for fertilizers and 3. the minimum sum of absolute value treatment that is an OR technique used to handle rather low quality Chinese historical data in which the least square method is unsuitable. This approach has been successfully implemented in China since 1980. The Chinese Academy of Sciences submits the annual grain output prediction report to the Chinese government in early May each year. The forward time of prediction is over half a year, and the average error rate of prediction is only 1.6 percent, with the right forecasting direction of bumper, normal or poor harvest. The achievement received the First Degree Award of Science & Technology Advancement of CAS in 1992 and the Third Degree Award of Science & Technology Advancement of China in 1996, and drew praise from top Chinese leaders, including former Premier Li Peng, Deputy Premier Zhou Jiahua and Li Lanqing. Xikang Chen is a professor and Xiaoming Pan is an associate professor at the Institute of Systems Science, Chinese Academy of Sciences. Their research involves input-output analysis and its application to grain output prediction (Chen) and environmental and resource accounting (Pan). The president of the OR Society of China, Xiang-Sun Zhang is professor and director of the Institute of Applied Mathematics, Chinese Academy of Sciences. His main research interests are in optimization theory and application, and artificial neural networks. The vice president of ORSC, Kan Cheng is a professor at the Institute of Applied Math. His main research interests are in reliability theory and stochastic optimization. Table of Contents OR/MS Today Home Page OR/MS Today copyright © 1998 by the Institute for Operations Research and the Management Sciences. All rights reserved. Lionheart Publishing, Inc. 506 Roswell Street, Suite 220, Marietta, GA 30060, USA Phone: 770-431-0867 | Fax: 770-432-6969 E-mail: lpi@lionhrtpub.com URL: http://www.lionhrtpub.com Web Site © Copyright 1998 by Lionheart Publishing, Inc. All rights reserved.