Graduation Date

Spring 2020

Document Type

Thesis

Program

Master of Science degree with a major in Natural Resources, option Forestry, Watershed, & Wildland Sciences

Committee Chair Name

Dr. Kevin Boston

Committee Chair Affiliation

HSU Faculty or Staff

Second Committee Member Name

Dr. James Graham

Second Committee Member Affiliation

HSU Faculty or Staff

Third Committee Member Name

Dr. Harold Zald

Third Committee Member Affiliation

HSU Faculty or Staff

Subject Categories

Forestry

Abstract

Timber transportation is one of the costliest activities for a forest company in Brazil and in many other countries, and it is a determining factor for the success of the forest enterprise. Thus, decision support tools are commonly used as methods to reduce these costs. The purpose of this study was to develop and analyze mathematical models to define the weekly timber transport schedule based on the monthly demands of the customers. The goal is to minimize the operational costs of forest transportation related to distances, timber freshness and road qualities. The decision process was made in two steps; the first was to select the timber location to be transported in a month, according to the client´s demand and timber stocks in the landing area. The second is to develop a weekly timber transportation scheduling to implement the monthly schedule. In the monthly decision process, three approaches in operational research were analyzed: multi-objective linear programming (MOLP) and two lexicographic multi-objective linear programming models (LMOLP 1 and LMOLP 2) with objectives in different hierarchical orders. The models were implemented in OPL (Optimization Programming Language) and its solution obtained using the software IBM ILOG CPLEX Optimization Studio. In the second part, the weekly timber scheduling, the decision process was taken to truck trips per week, ensuring timber transportation according to the customer's desired post-harvest age and a balance of truck trips per week. In this second stage, a Lexicographic Goal Programming model was developed due to a clear priority ordering amongst the goals to be achieved, in which in the sum of days left to deliver the timber from week 1 will be less than week 2; week 2 will be less than week 3; so on. The model was applied in the software Lindo.

The results obtained from the monthly decision-making process reveal that the flexibility of the lexicographic models demonstrate great potential for a reduction in costs. Total costs for the LMOLP 1 model were 30% less than the cost resulting from the MOLP model, and 9% less than the LMOLP -2 model. Regarding the second decision-making process, the lexicographic goal programming was highly suitable to solve weekly planning problems with complex multi-attribute nature.

Citation Style

APA

Share

 
COinS