Thesis (Ph. D.)--University OF MICHIGAN. / CHAIRMAN: THOMAS J. ARMSTRONG.

Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006. / Source: Dissertation Abstracts International, Volume: 67-11, Section: B, page: 6663. Adviser: Udatta S. Palekar. Includes bibliographical references (leaves 193-201) Available on microfilm from Pro Quest Information and Learning.

The purpose of this thesis is to analyze the business management process for small to medium size enterprises (SMEs) and to develop an automated product management system for integrated business processing. Due to the relatively small scale of operation of these companies, they experience difficulties in applying readily available tools in the area of product management. The target of the analysis is to develop a prototype software system for automatic product management. Moreover, this software developed for this thesis is applied to a real business case---that of Inshore Fisheries, a small scale integrated fishing operation in Atlantic Canada. The system consists of six components that are grouped into two categories of input and output analysis. The Input category includes production operations and customer ordering. The Output category includes four performance reports: (1) order invoices, (2) production worksheets, (3) inventory tracking, and (4) periodic income statements. The entire system is implemented using Visual Basic 6.0. The database is designed using Microsoft Access. The four performance reports in the output are designed using Microsoft Excel spreadsheets. Furthermore, a GUI integrates all components together to form a complete business processing system. Based on the results of the application of this system to Inshore Fisheries, this software solution can be applied to other SMEs by making specific modifications but without revising the basic principle and system design of the software system. This system possesses a great amount of flexibility and extendibility for practical business process decision-making.

With advent of complex and expensive automated manufacturing systems tool management is becoming an important issue both in design and implementation phases. The use of highly automated technology has resulted in an increasing need to integrate tool management more thoroughly into other planning and operational issues of these systems. In this research effort a discrete events simulation model has been developed to analyze the system performance under different operational strategies. These strategies include three tool configurations; identical, nonidentical and hybrid, and two part loading patterns; batch and random loading. In addition, a new tool refilling concept, tool replenishing, is also examined. The replenishing concept consists of two policies; preventive and failure replenishing. The effects of shared and dedicated tooling policies have also been investigated. With the comparison of performance measures for each set of operational strategies, the mutual advantages and concerns of each of them are highlighted. Within the simulation scope, the shared tooling policy is significantly superior to the dedicated tooling policy. It is also shown that identical tool configuration outperforms the other tool configurations. These have been confirmed by statistical tests performed on the simulation results.

The wide application of NC technologies and the advance in integrated manufacturing have significantly improved productivity. This however also leads to complexity in shop floor planning, particularly in machine-level planing due to the increased flexibility in the selection of machining parameters, tools, and tool paths. Part sequencing, tool replacement and machining speed selection for metal cutting in general and tool set selection, machining speed specifications as well as path sequencing for hole making in particular have a direct impact on manufacturing economics and hence have drawn much attention of many researchers. However, these problems have been often solved in isolation to each other, thereby causing inconsistent and conflicting planning actions on the shop floor. As a result, the solutions obtained in such a way can not be used for real shop floor planning. At the best a lengthy process is needed to resolve the conflict between the separately obtained solutions. Such a decision process obviously does not meet the need of modern manufacturing environment where quick and consistent planning decisions are imperative. The main purpose of this study is to model and solve several combined planning problems facing today's manufacturing industry. These include (a) part sequencing and tool replacement with sequence-dependent setup times and probabilistic tool life; (b) Just-In-Time (JIT) part scheduling with variable processing times and sequence dependent setups; and (c) tool set selection, machining speed specification, operation sequencing and path selection for hole making operations. These problems are combinatorial in nature and are often classified as NP-complete. Consequently, optimal solutions may not be obtained within polynomial times. In this dissertation, tabu search technique has been employed to solve the above combined planning problems. The computational results have shown that these problems can be efficiently solved and consistent decisions can be made based on the solutions. The effects of some important parameters such as initial solutions, move selection, termination criteria, and tabu list size on the search performance have also been examined.

The present study is concerned with nonlinear analysis of toroidal shells and storage tanks. The study serves to demonstrate the usefulness of the differential quadrature method in this area of computational mechanics. The problem of the response of axisymmetric toroidal shells to uniform external pressure forms the first part of this study. Nonlinear thin shell theory, accounting for large displacements, is employed. The new differential quadrature method is used to obtain numerical results. For validation a bifurcation solution is found using the finite element method. The commercial code ADINA is used for this purpose. Finally the two methods are used to provide results for eight cases of toroidal shells. An axisymmetric analysis of a liquid storage tank with a circular base plate resting on an elastic foundation forms the second part of this study. Nonlinear shell theory is used for the tank wall and base plate, while a linear model is used for the foundation. A convergence study is carried out to determine the appropriate analysis parameters for the method. Partial validation is obtained by comparison with previously published results. These results are compared with finite element method results. Additional results are presented covering a wide range of tank geometric parameters.

Flexible Manufacturing Cells (FMC) may be considered the most significant development in small batch manufacturing. Setting up and operating costs prove to be the most major hindrance to large scale implementation and use of the FMCs, particularly by small and medium size industries. Incompatibility between the different components constituting the cells, and the lack of a unified language/approach to programming and coordinating them, are cited as the cause of the complexity of setting up and subsequently operating the cells. In order to eliminate these difficulties, a new philosophy for setting up, programming, and control of FMCs has been developed. This thesis reports part of the effort for developing the new "UniSet" philosophy and the components of the UniSet Environment. UniSet has been developed as a Unified Manufacturing Instruction Set based on a comparison of machine tool and programming primitives. UniSet allows programmers, if they so desire, to concentrate on only one instruction set rather than numerous machine programming languages. Task level UniSet has been designed using an object-oriented and a knowledge based approach to eliminate cell programming difficulties. An expert cell programmer can be replaced by a less experienced manufacturing operator. A software environment for Computer Integrated Manufacturing (CIM), called UniSet Environment, has been developed to address the shortcomings of setting up and operating FMCs comprising incompatible components from the different suppliers. The UniSet Environment provides the users with a consistent platform to configure, program, simulate, and control an FMC irrespective of the nature of its constituents. The Environment is coded an Object-Oriented Programming (OOP) language, Smalltalk. The Environment consists of four modules which capture complete information about the cell program, existing cell facilities and databases, and distribute the information into the corresponding Object Models. Each of the module is responsible for a specific function. The Setup module is primarily used for defining the cell, its components and their functionalities, and reconfiguring a cell or building preparation databases. The Programming module allows the user to edit cell programs which may be developed in UniSet, Task level UniSet, or native machine languages. The programs written in Task level UniSet are used to generate detailed UniSet codes which are translated into specific machine codes by the generation and translation module. The Simulation/Compilation module is used for generating the control structure--Task Initiation Diagram (TID)--for the operation of the cell according to the user program as well as knowledge contained in the databases. The Control module is used to coordinate the workstation, as well as carry out the task of harmonizing between the different communication protocols of the devices.

The production efficiency has been significantly improved over the years due to the rapid advancement in CAD and CAM. Very often, however, the production cycle is still unacceptably long. This has been frequently caused by the bottleneck link between CAD and CAM. The CAD design model cannot be directly machined by a CNC machine without an NC program, an NC program cannot be coded without a clearly defined tool path, and the CAD data are not readily useable for preparing tool path. A large amount of time is often spent on data exchange, NC tool path generation, and NC program preparation. To further reduce the production cycle, the CAD model has to be documented in a neutral format such that it can be readily used in CAM and the tool path should be automatically generated from the CAD model in a neutral format. The recently developed STEP data exchange standard provides such a neutral format that is independent of any software and hardware. However, none of the existing tool path generation systems is based on STEP. In view of the above facts, a STEP based tool path generation system is developed for rough machining of planar surfaces. This system comprises four modules. The first one is the data extraction module used to obtain geometric data from STEP. The second module is the slicing module which is designed to convert the three dimensional tool path generation problem into a series of two dimensional problems. The tool path will be generated by the third module. The tool path is then translated into NC program by an NC code generator--the last module. The proposed system has been implemented in a UNIGRAPHICS CAD/CAM environment. The application of the system is illustrated using an example part.

The concept of Flexible Manufacturing Cells (FMCs) proves to be one of the major productivity enhancement tools for batch processing. The cost and expertise required to setup a cell, and subsequently to operate it, negates the many advantages of implementing FMCs. This problem hampers the introduction of the concept in many industries. At present, combining machines from different manufacturers to form an FMC is cumbersome. Incompatibility due to unique characteristics and varying programming and control languages among machines present the largest difficulties. This thesis demonstrates that it is possible to create a manufacturing instruction set and environment labelled UniSet (the acronym for $\rm\underline{Uni}$fied instruction $\rm\underline{Set}$) that enables programming and controlling a flexible manufacturing cell with a set of common instructions. UniSet is a software environment designed to provide the FMC user with a consistent platform in which to configure, program, simulate and control an FMC irrespective of the constituent machines. The environment is coded in an Object Oriented Programming (OOP) language; Smalltalk. This thesis describes UniSet as an instruction set, the concept behind the UniSet programming environment and the initial developments of the environment. (Abstract shortened by UMI.)

A complete computational model for industrial robot programming is presented. There are two main objectives in realizing this model. The first is to allow shop floor production engineers (application programmers) to create and modify sensor-based robot programs. The proposed iconic user interface provides a non-textual programming mechanism. The icons, which represent individual robot skills, are linked and parameterized to modify the behaviour of the skills. Use of a control flow mechanism, as opposed to data flow, makes the description of the robot operation as a set of skills immediately obvious. Linking the skill icons requires only a few control constructs which makes the interface usable on the shop floor. This system provides a mechanism for online creation and debugging of sensor-based robot operations. The second objective is to enable the system programmer to create and maintain the robot skills using consistent and facilitated methods. This is the underlying software architecture that makes the iconic shop floor interface possible. It is an object-based method that provides functional abstraction of the sensors and machines. The objects include skills, sensor drivers, logical sensors, and machine drivers. The skills are defined in the form of templates that completely specify a sensor-based robot action. Other significant results ensue from the two listed above, for example, the possibility of standardization in robot programming at the skill level. The ability to separate the responsibilities of individuals with different capabilities is another objective that has the side effect of making robot systems development manageable. The computational model presented is called "SKills-Oriented Robot Programming (SKORP)." In this model the skills execute exclusive of each other and therefore the computation for each skill can be represented independently. Skills are designed and documented using realtime design tools from the multiactivity paradigm. The SKORP model provides consistent and usable design methods for describing computation in embedded systems. These design tools are used by the system programmer to guarantee the realtime interaction of the software modules that compose a skill. This research is directed toward industrial robotics in traditional and non-traditional habitats, but the model presented is equally applicable to any numerically controlled machine that either requires sensors or interacts with the environment in a complex way.