Quantitative measure of sustainability is a hot topic in the fields of sustainable development, economics, ecology and relative disciplines. An evolution model put forward by Ulanowicz is introduced here to evaluate the trend of economic sustainability. The model, which is based on thermodynamic, network analysis, information theory and researches on real-life ecosystems, could measure sustainability of any complex, matter/energy flow system. It says that the capacity for a system to undergo evolutionary change or self-organization consists of two factors: 1) ascendency: the network′s capacity to perform in a sufficiently organized and efficient manner as to maintain its integrity over time; and 2) resilience: its reserve of flexible fall-back positions and diversity of actions that can be used to meet the exigencies of novel disturbances and the novelty needed for on-going development and evolution. These two factors are complementary with respect to diversity and connectivity in the network. A system′s resilience is enhanced by more diversity and connectivity, while ascendency is increased by less diversity and connectivity. In other words, too much ascendency (resilience) means too little resilience (ascendency). The key point is that system′s sustainability is decided by the balance between ascendency and resilience. A System with too little ascendency has neither the extent of activity nor the internal organization needed to survive. Conversely, systems with too much ascendency appear brittle in the face of novel disturbances. In one word, system′s sustainability is depending on the tradeoff between ascendency and resilience. In this paper, the trend of economic sustainability in Beijing from 1985 to 2005 was evaluated. Firstly, based on I-O table cash flows among six fundamental sectors of Beijing economy were outlined: agriculture; industry; construction; transportation, post and telecommunication; commerce and catering trade; other services. Secondly, the value of ascendency/development capacity (denoted as a), which is a relative measure of the organized power flowing within the system, was calculated. When the system is at its optimal sustainability a works out to be aopt. The further the a is from aopt, the less sustainable the system is. In this paper we assumed that aopt is 0.37. The results showed that during 1985-1987 where a was arising from 10.94% to 11.61% system developed toward configuration of more sustainable. While during 1988-1992 where a was dropping from 11.61% to 9.66% system developed toward configuration of less sustainable. During 1993-2002 system developed toward the optimal sustainability. While during 2003-2005 system kept away from the optimal state again. Third, the way to vector the system towards configuration of more sustainability was analyzed. During 1985-2005, each a is less than 0.37, which means that the system inhabits much more resilience than ascendency. Sensitivity analysis on ascendency with respect to each individual component process quantified the value of that link “at the margin” in 2005. The analysis revealed that to strengthen paths contributing more to ascendency is an effective way to increase system′s sustainability. For example, when artificially doubled the flow of construction′s final demand the value of a was increasing from 9.66% to 10.42%.