David J. Singer

Multiobjective Particle Swarm Optimization of a Planing Craft with Uncertainty

Uncertainty exists in many of the design variables and system parameters for planing craft. This is especially true in the early stages of design. For this reason, and others, optimization of a craft’s performance characteristics is often delayed until later in the design process, after uncertainties have been at least partially resolved. However, delaying optimization can also limit its potential, because freedom to make changes to a design is also highly limited in the later stages. This paper demonstrates how uncertainty can be directly incorporated into optimization using particle swarm. A simple synthesis model for a planing craft is built, and a deterministic Pareto front of optimal solutions is found, minimizing two objectives; drag and vertical acceleration at the center of gravity. The craft’s weight is then modelled as a normally distributed random variable and sampling methods are used to quantify the uncertainty in the estimated drag for points along the Pareto front. Preliminary results reveal that drag uncertainty is not constant along the Pareto front, presenting useful trade-off information for designers and decisions makers.

Dynamic control of a closed two-stage queueing network for outfitting process in shipbuilding

Abstract The U.S. Naval shipbuilding industry faces significant challenges to build ships on-time and within budgeted cost. To achieve greater efficiency and timeliness in shipbuilding, we developed a flexible two-stage queueing model under a CONWIP job release policy to enhance the planning and control of the outfitting process, one of the key processes in shipbuilding. The model is formulated using Markov Decision Processes which can provide (1) the optimal dynamic control policy and (2) the optimal cost. The numerical results showed that the optimal control policy is a state dependent threshold type policy and very complex to analyze. Therefore, we developed a static model to simplify the dynamic model and used Mean Value Analysis to gain insights. Using both data from the dynamic model and the static model, we developed a regression model to calculate a threshold policy heuristic. Testing reveals that the performance of this heuristic is very close to the optimal.

Analytical approach to a two-stage queuing network for the planning of outfitting processes in shipbuilding

As an important part of shipbuilding, outfitting can represent a major portion of the material cost as well as the construction time of a vessel. However, as a result of disturbances caused by unexpected delays, system variations, and technological constraints, scheduling of outfitting processes is therefore complex and can lead to major construction delays. To improve the shipbuilding system efficiency in the presence of variation, a two-stage strategic-level outfitting planning model has been developed. The results of the model provide the optimal percentage of outfitting work that should be completed at each stage given the introduction of variation. The model presented can be used as a benchmark for current ship-outfitting management decisions.

Dynamic control of the N queueing network with application to shipbuilding

The US shipbuilding industry faces challenges of building ships on time and within budgeted cost. We introduce an operational flexibility to shipbuilding to improve the system control. We model the flexible ship production system as an ‘N’ queueing network. However, the ‘N’ network model still lacks effective and computationally lightweight policies, especially with non-preemption. We use a Markov Decision Process (MDP) to gain structural insights into the optimal control policy. We develop a state dependent Optimal Threshold policy and benchmark it against other policies to show its excellent robustness and effectiveness. Our extensive test suite shows that (1) the Optimal Threshold policy performs the best in all the heuristics we tested; and (2) the cost of the system under control of this threshold policy is very close to the optimal cost calculated by the MDP. To calculate the exact optimal threshold level is difficult; therefore, we develop a birth-death process to determine a Analytical threshold level. Based on the optimal threshold values over a large test suite, we refine the analytical threshold level using a second-order regression model. We find the performance of the Regression Threshold policy to be within a few percent of optimal.

Including Principles of Set-Based Design in Multidisciplinary Design Optimization

The traditional point-based design process is highly iterative and can be inefficient, particularly for for multidisciplinary problems. Set-based design constitutes a design space reduction process that offers improvements over the point-based approach. In set-based design, designers begin with a broad set of values for the design variables, then gradually narrow the sets as more information becomes available. The principles of set-based design are reflected in the development of a new multidisciplinary design optimization algorithm. The algorithm returns the optimal choice for the reduced design space, instead of a single specific value for each design variable. The new multidisciplinary design optimization algorithm inspired by set-based design was used in a multidisciplinary ship design application.

Prospect Theory-based Real Options Analysis for Noncommercial Assets

When an engineering system has the ability to change or adapt based on a future choice, then flexibility can become an important component of that system’s total value. However, evaluating noncommercial flexible systems, like those in the defense sector, presents many challenges because of their dynamic nature. Designers intuitively understand the importance of flexibility to hedge against uncertainties. In the naval domain, however, they often do not have the tools needed for analysis. Thus, decisions often rely on engineering experience. As the dynamic nature of missions and new technological opportunities push the limits of current experience, a more rigorous approach is needed. This paper describes a novel framework for evaluating flexibility in noncommercial engineering systems called prospect theory-based real options analysis (PB-ROA). While this research is motivated by the unique needs of the U.S. Navy ship design community, the framework abstracts the principles of real options analysis to suit noncommercial assets that do not generate cash flows. One contribution of PB-ROA is a systematic method for adjusting agent decisions according to their risk tolerances. The paper demonstrates how the potential for loss can dramatically affect decision making through a simplified case study of a multimission variant of a theoretical high-speed connector vessel.

Multidisciplinary Design Optimisation of a Ship Hull Using Metamodels

Abstract In this paper, the metamodelling approach is applied to the multidisciplinary design optimisation of a ship hull with regard to resistance, seakeeping and maneuvering performance. The hull shape is described by a set of design variables used in the simulations. At the top system level, a simple cost metric is defined to drive the overall design optimisation process. Changes to the hull shape are reflected in the numerical model for resistance computations and in the seakeeping and maneuvering assessment. An automated process has been developed for propagating changes to the hull form in the numerical model used for the resistance computations; this expedites the computations at the sample points used for developing the metamodels. The validity of employing metamodels instead of the actual analysis methods during the optimisation is demonstrated by comparing the values of the objective functions and constraints at the optimum point when using the actual analysis methods and the metamodels. The effectiveness of the multidisciplinary design optimisation algorithm is demonstrated using a simple analytical example.

Effects of uncertainty in fuzzy utility values on general arrangement optimisation

Abstract The Intelligent Ship Arrangements (ISA) software system assists designers in developing rationally-based arrangements, which should satisfy design specific needs, general Navy requirements and standard practices. The system is intended to be used following Advanced Ship and Submarine Evaluation Tool (ASSET). The optimisation core uses fuzzy programming to provide the framework upon which the objective function and constraints are based. This methodology relies on fuzzy utility functions that quantify the satisfaction of design goals and constraints, which allows for a very powerful and generic foundation for problem description. The uncertainty of the fuzzy utility preference values and the corresponding impacts on design optimisation are studied. The uncertainty is introduced by inserting a normal distributed stochastic perturbation into the fuzzy utility values.