Creating a recycling network requires the use of appropriate decision support tools. Decisions regarding the location of entities should take into account as many factors as possible, covering technical, economic, environmental, and legal issues. Thanks to this, shaping a selected part of the network, e.g., in the context of its expansion and the location of new entities, will ensure the maximization of benefits both from the point of view of network participants, as well as vehicle owners and other interested parties.Searching for a solution to the decision-making problem of locating elements of a recycling network structure requires defining a criterion function, which will serve as the deciding criterion for which of the permissible decisions is the optimal one. This function is therefore an indicator for evaluating the quality of the solution.
The optimal solution is one for which the adopted quality evaluation criterion (e.g., minimum cost) reaches an extreme value while satisfying [6]:
− network constraints (e.g., processing capacity), − conditions resulting from the demand for waste processing (e.g., number of vehicles and mass of waste derived from them to be managed), − conditions resulting from the physical interpretation of quantities (e.g., non-negativity).
The optimization criterion is a reflection of the preference function of the decision-maker.
Depending on the number of preference functions included as partial criterion functions, optimization tasks can be divided into single-criterion and multi-criteria [6,7,10,11]. Single-criterion tasks are those in which one criterion determines the choice of the optimal variant.
In multi-criteria tasks, several preference functions are considered simultaneously, often reflecting opposing views. In models of this type, we can have two or more partial functions that are extremized simultaneously, where the optimal realization of one goal most often leads to a deterioration in the realization of at least one of the other goals.
The decision is therefore made in the context of the simultaneous realization of all distinguished partial criterion functions, although as a rule, there is no single decision (solution, action) that is the best from all points of view simultaneously. Unlike classical operations research techniques, multi-criteria methods do not provide objectively best solutions. (…)
The article contains 30,682 characters.
Source: Logistyka Magazine 4/2012
Optimization criteria in the location of vehicle recycling network entities – PDF file for download







