一种改进的多学习教与学优化算法

针对传统教与学算法在解决复杂多峰函数优化问题时,具有局部最优且搜索开发能力较差的缺点,提出了一种改进的多学习教与学优化算法,新算法为学员的每一维加入不同的教学因子,设计了基于学员均值比较的教师选择策略和向教师及学员学习的多学习策略。基于多个单峰和多峰函数的仿真结果表明,新算法跟传统的、改进的教与学算法相比,在稳定性、寻优精度和收敛速度方面更具优势。     

Spatial based compromise programming for multiple criteria decision making in land use planning

Today, competing land use is continuing to occur in many developed regions. In the Agricultural Development Zone of Western Sydney Region, which is characterised by complex landscape patterns, land use competition is widespread. From a land use planning perspective, identification of suitable locations for a given type of land use is necessary for decision makers to formulate land use alternatives in different locations, based on existing land potential and constraints. For such a region, use of a simple method that implements a categorical system and considers only inherent land characteristics in the analysis is often inadequate to arrive at an optimal spatial decision. The primary aim of this paper is to develop spatial modelling procedures for agricultural land suitability analysis using compromise programming (CoPr) and fuzzy set approach within a geographical information systems (GIS) environment. Five main sets of spatial data for use as decision criteria were developed by using fuzzy set methodology: a land suitability index (LSI) for maximising the land productivity objective; an erosion tolerance index (ETI) for minimising the erosion risk objective; a runoff curve number (CN) for maximising the water discharge regulation objective; an accessibility (RP) measure for maximising the land accessibility objective; and the proximity to water body (WP) for minimising the water pollution objective. An L _p -metric was used in the analysis utilising different strategies with representative indices ranging from a situation where full tradeoff among criteria occurs to a noncompensatory condition. Different weighting combinations were also applied, and decision analysis was carried out by using values ranging from 0 to 1.0, where 1.0 is considered as an ideal point. The CoPr model demonstrated in this paper yielded a promising result, as several different techniques of sensitivity analysis show reasonably good results. Likewise, an overlay of that result with the present land use/land cover indicates a good corresponding spatial matching between existing land use (orchard and cultivated land), and the cells (land parcels) classified as the best in CoPr. The results are amenable to various map display techniques, either using continuous values or by defining different cut off points in the data space within a raster GIS environment.     

Aerial surveys of manatees (<Emphasis Type="Italic">Trichechus manatus)</Emphasis> in Lee County,Florida<Emphasis Type="Italic">,</Emphasis> provide insights regarding manatee abundance and real time information for managers and enforcement officers

Conservation and management of the endangered Florida manatee is often centered on reducing mortality caused by watercraft collisions. Lee County, Florida, has led the state in watercraft-related mortality for eight of the last 10 years. This county is of particular concern as it contains important habitat for manatees, including extensive feeding grounds and an artificial warm-water refuge where more than 900 manatees have been recorded on a single day. Distributional aerial surveys were conducted from April 2007 through April 2009 over Lee County waters. Surveys yielded higher numbers of manatees than previously observed in this area. Using GIS methodology, kernel density analysis illustrated seasonal changes in distribution patterns and highlighted areas where manatees were most densely clustered. For example, during summer months, manatees were widely distributed throughout the survey area, with high-density areas associated with seagrass beds. During winter months, manatees were densely clustered at warm-water sites and over feeding grounds within close distance of these sites. These seasonal distribution patterns coincide well with speed zone designations. Counts and distributions of manatees were made available, almost immediately if necessary, to local marine law enforcement in an attempt to focus resources toward reducing manatee-watercraft collisions. Future studies should implement similar communication strategies to improve conservation efforts.     

Modelling cheetah relocation success in southern Africa using an Iterative Bayesian Network Development Cycle

Relocation is one of the strategies used by conservationists to deal with problem cheetahs in southern Africa. The success of a relocation event and the factors that influence it within the broader context of long-term viability of wild cheetah metapopulations was the focus of a Bayesian Network (BN) modelling workshop in South Africa. Using a new heuristics, Iterative Bayesian Network Development Cycle (IBNDC), described in this paper, several networks were formulated to distinguish between the unique relocation experiences and conditions in Botswana and South Africa. There were many common underlying factors, despite the disparate relocation strategies and sites in the two countries. The benefit of relocation BNs goes beyond the identification and quantification of the factors influencing the success of relocations and population viability. They equip conservationists with a powerful communication tool in their negotiations with land and livestock owners, which is key to the long-term survival of cheetahs in southern Africa. Importantly, the IBNDC provides the ecological modeller with a methodological process that combines several BN design frameworks to facilitate the development of a BN in a multi-expert and multi-field domain.     

Combating desertification in Iran over the last 50 years: An overview of changing approaches

Desertification in Iran was recognized between the 1930s and 1960s. This paper traces Iran’s attempts to reclaim desertified areas, evaluates the anti-desertification approaches adopted, and identifies continuing challenges. Iran has areas vulnerable to desertification due to extensive areas of drylands and increasing population pressure on land and water resources. Over-grazing of rangelands is a particular problem. Initially desertification was combated mainly at the local level and involved dune stabilization measures, especially the use of oil mulch, re-vegetation and windbreaks. Insufficient technical planning in the early years has led to changed approaches to plant densities and species diversity in plantations, and increased on-going management of existing plantations. Since the late 1980s forage and crop production has increased in areas where runoff control techniques are practiced. The social and economic aspects of anti-desertification programs have assisted in poverty reduction by providing off-season employment in rural areas. In 2004 a national plan to combat desertification was ratified and this placed an emphasis on community participation. Continuing challenges include managing existing desertified areas as well as taking into account potential future problems associated with rapidly depleting groundwater supplies and a predicted reduction in the plant growth period accompanying climate change.     

A multi-criteria decision analysis assessment of waste paper management options

The use of Multi-criteria Decision Analysis (MCDA) was investigated in an exercise using a panel of local residents and stakeholders to assess the options for managing waste paper on the Isle of Wight. Seven recycling, recovery and disposal options were considered by the panel who evaluated each option against seven environmental, financial and social criteria. The panel preferred options where the waste was managed on the island with gasification and recycling achieving the highest scores. Exporting the waste to the English mainland for incineration or landfill proved to be the least preferred options. This research has demonstrated that MCDA is an effective way of involving community groups in waste management decision making.