Climate change impacts and adaptation in Bangladesh : an agent-based approach

Bangladesh exemplifies the complex challenges facing densely populated coastal regions. The
pressures on the country are immense: around 145 million people live within an area of just 145,000 sq-km at
the confluence of three major river systems: the Ganges, the Brahmaputra and the Meghna. While progress
has been made, poverty remains widespread, with around 39% of children under five malnourished. Most of
its land-mass lies below 10m above sea level with considerable areas at sea level, leading to frequent and
prolonged flooding during the monsoons. Sea level rise is leading to more flooding as storm surges rise off
higher sea levels, pushing further inland. Higher sea levels also result in salt-water intrusion into freshwater
coastal aquifers and estuaries, contaminating drinking water and farmland. Warmer ocean waters are also
expected to lead to an increase in the intensity of tropical storms.
Bangladesh depends on the South Asian summer monsoon for most of its rainfall which is expected to
increase, leading to more flooding. Climate scientists are also concerned about the stability of monsoon and
the potential for it to undergo a nonlinear phase shift to a drier regime. Bangladesh faces an additional
hydrological challenge in that the Ganges and Brahmaputra rivers both rise in the Himalaya-Tibetan Plateau
region, where glaciers are melting rapidly. The Intergovernmental Panel on Climate Change (IPCC)
concluded that rapid melting is expected to increase river flows until around the late-2030s, by which time
the glaciers are expected to have shrunk from their 1995 extent of 500,000 sq-km to an expected 100,000 sqkm.
After the 2030s, river flows could drop dramatically, turning the great glacier-fed rivers of Asia into
seasonal monsoon-fed rivers. The IPCC concluded that as a result, water shortages in Asia could affect more
than a billion people by the 2050s. Over the same period, crop yields are expected to decline by up to 30% in
South Asia due to a combination of drought and crop heat stress. Bangladesh is therefore likely to face
substantial challenges in the coming decades.
In order to adequately understand the complex, dynamic, spatial and nonlinear challenges facing Bangladesh,
an integrated model of the system is required. An agent-based model (ABM) permits the dynamic
interactions of the economic, social, political, geographic, environmental and epidemiological dimensions of
climate change impacts and adaptation policies to be integrated via a modular approach. Integrating these
dimensions, including nonlinear threshold events such as mass migrations, or the outbreak of conflicts or
epidemics, is possible to a far greater degree with an ABM than with most other approaches.
We are developing a prototype ABM, implemented in Netlogo, to examine the dynamic impacts on poverty,
migration, mortality and conflict from climate change in Bangladesh from 2001 to 2100. The model employs
GIS and sub-district level census and economic data and a coarse-graining methodology to allow model
statistics to be generated on a national scale from local dynamic interactions. This approach allows a more
realistic treatment of distributed spatial events and heterogeneity across the country. The aim is not to
generate precise predictions of Bangladesh’s evolution, but to develop a framework that can be used for
integrated scenario exploration. This paper represents an initial report on progress on this project. So far the
prototype model has demonstrated the desirability and feasibility of integrating the different dimensions of
the complex adaptive system and, once completed, is intended to be used as the basis for a more detailed
policy-oriented model.

An agent-based approach to value theory and wealth distribution in economics

Any attempt to model an economy requires foundational assumptions about the relations between prices, values and the distribution of wealth. These assumptions exert a profound influence over the results of any model. Unfortunately, there are few areas in economics as vexed as the theory of value. I argue in this paper that the fundamental problem with past theories of value is that it is simply not possible to model the determination of value, the formation of prices and the distribution of income in a real economy with analytic mathematical models. All such attempts leave out crucial processes or make unrealistic assumptions which significantly affect the results. There have been two primary approaches to the theory of value. The first, associated with classical economists such as Ricardo and Marx were substance theories of value, which view value as a substance inherent in an object and which is conserved in exchange. For Marxists, the value of a commodity derives solely from the value of the labour power used to produce it - and therefore any profit is due to the exploitation of the workers. The labour theory of value has been discredited because of its assumption that labour was the only ‘factor’ that contributed to the creation of value, and because of its fundamentally circular argument. Neoclassical theorists argued that price was identical with value and was determined purely by the interaction of supply and demand. Value then, was completely subjective. Returns to labour (wages) and capital (profits) were determined solely by their marginal contribution to production, so that each factor received its just reward by definition. Problems with the neoclassical approach include assumptions concerning representative agents, perfect competition, perfect and costless information and contract enforcement, complete markets for credit and risk, aggregate production functions and infinite, smooth substitution between factors, distribution according to marginal products, firms always on the production possibility frontier and firms’ pricing decisions, ignoring money and credit, and perfectly rational agents with infinite computational capacity. Two critical areas include firstly, the underappreciated Sonnenschein-Mantel- Debreu results which showed that the foundational assumptions of the Walrasian general-equilibrium model imply arbitrary excess demand functions and therefore arbitrary equilibrium price sets. Secondly, in real economies, there is no equilibrium, only continuous change. Equilibrium is never reached because of constant changes in preferences and tastes; technological and organisational innovations; discoveries of new resources and new markets; inaccurate and evolving expectations of businesses, consumers, governments and speculators; changing demand for credit; the entry and exit of firms; the birth, learning, and death of citizens; changes in laws and government policies; imperfect information; generalized increasing returns to scale; random acts of impulse; weather and climate events; changes in disease patterns, and so on. The problem is not the use of mathematical modelling, but the kind of mathematical modelling used. Agent-based models (ABMs), objectoriented programming and greatly increased computer power however, are opening up a new frontier. Here a dynamic bargaining ABM is outlined as a basis for an alternative theory of value. A large but finite number of heterogeneous commodities and agents with differing degrees of market power are set in a spatial network. Returns to buyers and sellers are decided at each step in the value chain, and in each factor market, through the process of bargaining. Market power and its potential abuse against the poor and vulnerable are fundamental to how the bargaining dynamics play out. Ethics therefore lie at the very heart of economic analysis, the determination of prices and the distribution of wealth. The neoclassicals are right then that price is the enumeration of value at a particular time and place, but wrong to downplay the critical roles of bargaining, power and ethics in determining those same prices.

Trade and industrial policy in developing countries : what scope for interfaces between agent-based models and computable-general equilibrium models

How best to assess trade and industrial policy in developing countries is a controversial question that unlocks a host of modelling complexities. Large computable general-equilibrium (CGE) models dominate many economic policy debates, but recent developments in the field have demonstrated that it is by no means clear that they give reliable results to questions of how trade reforms affect the poor. Over the last decade or so, a new approach to modelling complex systems has emerged using agent-based models (ABMs). This paper explores the question of whether ABMs are useful for economic policy-makers seeking to quantitatively model the effects of trade and industrial policies and whether constructive interfaces could be developed between CGE models and ABMs. The paper argues that in developing economic policy, ABMs can and should be used in conjunction with CGE models and that there is much to be gained from a greater understanding of the strengths and weaknesses of different modelling approaches, and what domains are most appropriate for their use. It concludes with some reflections on the reasons for the success of CGE approaches and ways in which ABMs could be made more widely understood and used among economists.

CARSA : A context-aware reasoning-based service agent model for AI planning of web service compostion

In order to achieve automatic and more intelligent service composition, dynamic description logic (DDL) is proposed and utilized as one emerging logic-level solution. However, reasoning optimization and utilization in such DDL-related solutions is still an open problem. In this paper, we propose the context-aware reasoning-based service agent model (CARSA) which exploits the relationships among different service consumers and providers, together with the corresponding optimization approach to strengthen the effectiveness of Web service composition. Through the model, two reasoning optimization methods are proposed based on the substitute relationship and the dependency relationship, respectively, so irrelevant actions can be filtered out of the reasoning space before the DDL reasoning process is carried out. The case study and experimental analysis demonstrates the capability of the proposed approach.

Adaptive data based neural network leader-follower control of multi-agent networks

In this paper, we propose a data based neural network leader-follower control for multi-agent networks where each agent is described by a class of high-order uncertain nonlinear systems with input perturbation. The control laws are developed using multiple-surface sliding control technique. In particular, novel set of sliding variables are proposed to guarantee leader-follower consensus on the sliding surfaces. Novel switching is proposed to overcome the unavailability of instantaneous control output from the neighbor. By utilizing RBF neural network and Fourier series to approximate the unknown functions, leader-follower consensus can be reached, under the condition that the dynamic equations of all agents are unknown. An O(n) data based algorithm is developed, using only the network’s measurable input/output data to generate the distributed virtual control laws. Simulation results demonstrate the effectiveness of the approach.

Mobile agent tracking with single frequency continuous wave radar : a linear robust filtering based approach

3-alkylpyrrole to the fabric surface. Direct applications of a conductive paint to the textile surface eliminate the exposure of the substrate to damaging oxidizing agents which allow the coating of more sensitive and delicate substrates. All textiles produced are tested for abrasion resistance and conductivity. For alkyl polypyrrole coated fabrics, the optimum carbon chain lengths are between n=10 and n=14, which result in optimum values of conductivity and solubility. The darkness of the tone is inversely related to the surface resistivity of the resulting conductive fabric. Therefore, deep black coatings have low resistivity whereas light gray coatings on a white fabric surface have higher surface resistivity. Longer alkyl chains result in higher surface resistivity in fabrics. The conductive coating of poly(3-decanylpyrrole) on the textile surface has a better abrasion resistance compared to that of an unsubstituted polypyrrole coating.

Learning other agents' preferences in multi-agent negotiation using the bayesian classifier

In multi-agent systems, most of the time, an agent does not have complete information about the preferences and decision making processes of other agents. This prevents even the cooperative agents from making coordinated choices, purely due to their ignorance of what others want. To overcome this problem, traditional coordination methods rely heavily on inter-agent communication, and thus become very inefficient when communication is costly or simply not desirable (e.g. to preserve privacy). In this paper, we propose the use of learning to complement communication in acquiring knowledge about other agents. We augment the communication-intensive negotiating agent architecture with a learning module, implemented as a Bayesian classifier. This allows our agents to incrementally update models of other agents' preferences from past negotiations with them. Based on these models, the agents can make sound predictions about others' preferences, thus reducing the need for communication in their future interactions.

A multi-agent incremental negotiation scheme for meetings scheduling

We present an agent-oriented approach to the meeting scheduling problem and propose an incremental negotiation scheme that makes use of a hierarchical structure of an individual agent's working knowledge. First, we formalise the meeting scheduling problem in a multi-agent context, then elaborate on the design of a common agent architecture of all agents in the system. As a result, each agent becomes a modularised computing unit yet possesses high autonomy and robust interface with other agents. The system reserves the meeting participants' privacy since there are no agents with dominant roles, and agents can communicate at an abstract level in their hierarchical structures