Stochastic Multi-criteria Analysis Research Articles

Port management and multiple-criteria decision making under uncertainty

The methodology for port management in this study takes into account the stochastic nature of climate as well as of use and exploitation agents. It provides a probabilistic characterization of the performance of different alternatives in terms of a set of indicators that reflect the potential benefits for different interest groups. The joint distribution functions thus obtained for a set of pre-defined management alternatives are used to compare strategies and to make decisions, based on stochastic multi-criteria analysis. This method, which also addresses the uncertainty of decision-maker preferences, ranked the strategies, and at the same time, specified the typical decision-maker preferences for each strategy. In the analysis of a hypothetical port, four alternatives were designed as solutions for port congestion. The criteria selected were the service quality provided to vessels, the total goods handled and the profits accrued by the Port Authority. The criteria measuring the performance of each strategy were statistically analyzed first. Then, assuming the uniform distribution of decision-maker preferences, the alternatives were ranked, and the relative importance of each criterion in the decision-making process was obtained.

Coastal zone management with stochastic multi-criteria analysis

The methodology for coastal management proposed in this study takes into account the physical processes of the coastal system and the stochastic nature of forcing agents. Simulation techniques are used to assess the uncertainty in the performance of a set of predefined management strategies based on different criteria representing the main concerns of interest groups. This statistical information as well as the distribution function that characterizes the uncertainty regarding the preferences of the decision makers is fed into a stochastic multi-criteria acceptability analysis that provides the probability of alternatives obtaining certain ranks and also calculates the preferences of a typical decision maker who supports an alternative. This methodology was applied as a management solution for Playa Granada in the Guadalfeo River Delta (Granada, Spain), where the construction of a dam in the river basin is causing severe erosion. The analysis of shoreline evolution took into account the coupled action of atmosphere, ocean, and land agents and their intrinsic stochastic character. This study considered five different management strategies. The criteria selected for the analysis were the economic benefits for three interest groups: (i) indirect beneficiaries of tourist activities; (ii) beach homeowners; and (iii) the administration. The strategies were ranked according to their effectiveness, and the relative importance given to each criterion was obtained.

Multi-criteria evaluation of multi-purpose stand treatment programmes for Finnish boreal forests under changing climate

Abstract The aim of this study is determine the potential of multiple-criteria analysis (MCA) methodology to assess the utility of different multi-objective stand treatment programmes, in the boreal forest. Simulations over a planning period of 100 years under current and changing climate scenarios by a physiological growth model provided input data for the stochastic multi-criteria analysis of forest management alternatives. Expected utility with regard to a set of objectives was estimated with an additive utility model. Timber production, carbon sequestration and biodiversity were considered as partial objectives; these in term were represented by decision criteria. The weights of partial objectives and criteria were subjected to uncertainty due to missing exact stakeholder preference information. Generic probability distributions for individual weights were defined based on qualitative information. Alternative stand treatment programmes with regard to the decision criteria were evaluated by means of linear preference functions. In a Monte Carlo analysis the utility model was employed to evaluate eight stand treatment programmes for their performance regarding the management objectives. The effect of climate change on utility of stand treatment programmes was examined also as an important factor. Over two sites and three tree species just two out of eight stand treatment programmes analysed in this study were able to dominate the set of alternatives; namely the unthinned scenario and the highest increase in thinning thresholds scenario (UT(0, 0) and BT(+30, +30)) in which stocking was kept at higher level compared to the currently recommended guidelines for site and species. Reducing uncertainty in the utility model by assuming accurate weights for the decision criteria did not show large changes neither in expected utility nor in the ranking of alternatives. Ignoring necessary revisions of how the forest is to be managed in the future might potentially result in losses, not only for timber production but also for benefits such as carbon sequestration and other amenities including biodiversity.

Stand level analysis on the effects of management and climate change on the growth and timber yield with implications on carbon stocks in boreal forest ecosystem: a model based approach

The main aim of this study was to investigate based on process-based ecosystem model simulations how management and climate change affect the growth, timber yield and carbon stocks in boreal forest ecosystem (Papers I-II). In addition, it was studied how the management affects under varying climate the possibility to meet multi-purpose demands in regard to timber production, carbon sequestration and biodiversity (in terms of dead wood) based on a stochastic multi-criteria analysis (Paper IV). Simulations were carried out in Scots pine (Pinus sylvestris L.), Norway spruce (Picea abies Karst. L.) and silver birch (Betula pendula) stands growing in the southern and northern Finland. In addition to current climate, two different climate change scenarios and eight different stand treatment programmes (i.e. one unthinned and seven thinning regimes) were applied in simulations over 100 years. The stand treatment programmes differed from each other in the sense that the mean stocking varied in the stand over the rotation. The simulation results were also used as data for a multi-criteria analysis, in which expected utility was calculated with an additive utility model (Paper IV). In this context, it was also utilized a wood products model developed (Paper III), to estimate carbon stocks in wood products. The model based analyses indicated that in unthinned stands, the growth and carbon stock was on an average always higher compared to thinned stands regardless of tree species, sites and climate scenarios compared (Papers I-II). However, in regard to thinning regimes, the one in which the mean stocking was kept an average higher over a rotation compared to the current thinning guidelines, had also a higher growth and timber yield (Paper I). This increased also the mean carbon stock both in the forest ecosystem and in the harvested timber compared to the current thinning guidelines (Paper II). The climate change (regardless of scenarios) in itself enhanced growth, timber yield and carbon stocks in relative terms more in the north than in the south regardless of tree species. The utility was also maximized, regardless of site, tree species and climate scenario applied, by leaving the stand unthinned over a rotation or alternatively by keeping the stocking level higher (e.g. 30% higher) over the simulation compared to the current thinning guidelines (Paper IV). When accurate weights for criteria of timber production, carbon sequestration and dead wood were used to reduce uncertainty, expected utility did not change noticeably. As a conclusion, the selection of stand treatment programme was found to have an important role, regardless of tree species, sites and climate scenarios, on the timber production, but also on other benefits such as carbon sequestration and amount of dead wood in the forest ecosystem.