Why Integrate?

There is a tendency to look at computer based solutions as inherently superior. This is an assumption that an engineer cannot afford to entertain. Some of the factors that justify an integrated system are listed below.

· a large organization where interdepartmental communication is a problem

· the need to monitor processes

· Things to Avoid when making a decision for integration and automation,

- ignore impact on upstream and downstream operations

- allow the system to become the driving force in strategy

- believe the vendor will solve the problem

- base decisions solely on financials

- ignore employee input to the process

- try to implement all at once (if possible)

· Justification of integration and automation,

- consider "BIG" picture

- determine key problems that must be solved

- highlight areas that will be impacted in enterprise

- determine kind of flexibility needed

- determine what kind of integration to use

- look at FMS impacts

- consider implementation cost based on above

· Factors to consider in integration decision,

- volume of product

- previous experience of company with FMS

- product mix

- scheduling / production mixes

- extent of information system usage in organization (eg. MRP)

- use of CAD/CAM at the front end.

- availability of process planning and process data

* Process planning is only part of CIM, and cannot stand alone.

Why Automate?

· Why ? - In many cases there are valid reasons for assisting humans

- tedious work -- consistency required

- dangerous

- tasks are beyond normal human abilities (e.g., weight, time, size, etc)

- economics

· When?

Figure 1.1 - Automation Tradeoffs

· Advantages of Automated Manufacturing,

- improved work flow

- reduced handling

- simplification of production

- reduced lead time

- increased moral in workers (after a wise implementation)

- more responsive to quality, and other problems

- etc.

· Various measures of flexibility,

- Able to deal with slightly, or greatly mixed parts.

- Variations allowed in parts mix

- Routing flexibility to alternate machines

- Volume flexibility

- Design change flexibility

CAD/CAM?

· A common part of an integrated system

· In CAD we design product geometries, do analysis (also called CAE), and produce final documentation.

· In CAM, parts are planned for manufacturing (eg. generating NC code), and then manufactured with the aid of computers.

· CAD/CAM tends to provide solutions to existing problems. For example, analysis of a part under stress is much easier to do with FEM, than by equations, or by building prototypes.

· CAD/CAM systems are easy to mix with humans.

· This technology is proven, and has been a success for many companies.

· There is no `ONE WAY' of describing CAD/CAM. It is a collection of technologies which can be run independently, or connected. If connected they are commonly referred to as CIM

The Architecture of Integration

· integrated manufacturing systems are built with generic components such as,

- Computing Hardware

- Application Software

- Database Software

- Network Hardware

- Automated Machinery

· Typical applications found in an integrated environment include,

- Customer Order Entry

- Computer Aided Design (CAD) / Computer Aided Engineering (CAE)

- Computer Aided Process Planning (CAPP)

- Materials (e.g., MRP-II)

- Production Planning and Control (Scheduling)

- Shop Floor Control (e.g., FMS)

· The automated machines used include,

- NC machines

- PLCs

- Robotics

- Material Handling / Transport

- Machines

- Manual / Automated Assembly Cells

- Computers

- Controllers

- Software

- Networks

- Interfacing

- Monitoring equipment

· On the shop floor computers provide essential support in a workcell for,

- CNC - Computer Numerical Control

- DNC - Direct Numerical Control of all the machine tools in the FMS. Both CNC and DNC functions can be incorporated into a single FMS.

- Computer control of the materials handling system

- Monitoring - collection of production related data such as piece counts, tool changes, and machine utilization

- Supervisory control - functions related to production control, traffic control, tool control, and so on.

General Concepts

· Manufacturing requires computers for two functions,

- Information Processing - This is characterized by programs that can operate in a batch mode.

- Control - These programs must analyze sensory information, and control devices while observing time constraints.

· An integrated system is made up of Interfaced and Networked Computers. The general structure is hierarchical,

· The plant computers tend to drive the orders in the factory.

· The plant floor computers focus on departmental control. In particular,

- synchronization of processes.

- downloading data, programs, etc., for process control.

- analysis of results (e.g., inspection results).

· Process control computers are local to machines to control the specifics of the individual processes. Some of their attributes are,

- program storage and execution (e.g., NC Code),

- sensor analysis,

- actuator control,

- process modeling,

- observe time constraints (real time control).

· The diagram shows how the characteristics of the computers must change as different functions are handled.

· To perform information processing and control functions, each computer requires connections,

- Stand alone - No connections to other computers, often requires a user interface.

- Interfaced - Uses a single connection between two computers. This is characterized by serial interfaces such as RS-232 and RS-422.

- Networked - A single connection allows connections to more than one other computer. May also have shared files and databases.

· Types of common interfaces,

- RS-232 (and other RS standards) are usually run at speeds of 2400 to 9600 baud, but they are very dependable.

· Types of Common Networks,

- IEEE-488 connects a small number of computers (up to 32) at speeds from .5 Mbits/sec to 8 Mbits/sec. The devices must all be with a few meters of one another.

- Ethernet - connects a large number of computers (up to 1024) at speeds of up to 10 Mbits/sec., covering distances of km. These networks are LAN's, but bridges may be used to connect them to other LAN's to make a WAN.

· Types of Modern Computers,

- Mainframes - Used for a high throughput of data (from disks and programs). These are ideal for large business applications with multiple users, running many programs at once.

- Workstations (replacing Mini Computers) - have multiprocessing abilities of Mainframe, but are not suited to a limited number of users.

- Micro-processors, small computers with simple operating systems (like PC's with msdos) well suited to control. Most computerized machines use a micro-processor architecture.

· A Graphical Depiction of a Workstation Controller

What is it?

Linux is an open source operating system. It is open because users and developers can use the source code any way they want. This allows anyone to customize it, improve it and add desired features. As a result Linux is dynamic, evolving to respond to the desires and needs of the users. In contrast, closed operating systems are developed by a single corporation using static snapshots of market models and profit driven constraints.

Linux is free. This allows companies to use it without adding cost to products. It also allows people to trade it freely. And, with the profit motive gone, developers have a heightened sense of community interest. The Linux community has developed a tremendous spirit because of these core development concepts.

A (Brief) History

Linux has existed since the early 1990s [3], but it grew out of previous developments in computing [4]. It was originally developed to be a Unix clone that would run on low cost computer hardware. Unix was developed in the 1970s. Through the 1970s and early 1980s it was used on large computers in companies and universities. During this time many refinements and enhancements were made. By the mid 1980s Unix was being used on many lower priced computers. By the end of the 1980s most universities were making use of Unix computers in computer science and engineering programs. This created a wealth of graduates who understood what they could expect from a mature operating system. But, it also created a demand to be able to do high level work at home on low priced machines.

Early in the 1990s Linux started as a project to create a Unix clone that would run on a personal computer. This project gained momentum quickly and by the mid 1990s it was ready for users. The first groups to adopt it were hobbyists, academics and internet services. At this time the general public was generally unaware of Linux but by the end of the 1990s it was beginning to enter the public sphere. By 2000 it had entered the popular press, and it was cited as a major threat to at least one existing operating system vendor. Now, it is available off-the-shelf in software and book stores.

1970s - Unix developed at AT&T labs by Ken Thompson and Dennis Ritchie

1980s - Unix became popular on high end computers

- The Unix platform is refined and matures

- Some versions of Unix were available for PCs - most notably QNX

1990s - Linus Torvalds begins working on a free Unix clone for PCs in 1991

- Others join the project it gets the name `Linux'

- By 1993 Linux begins to enter the mainstream of computer users

- Linux machines constitute a large number of servers on the Internet

- Many large companies begin to support Linux - e.g. Dell, IBM

2000s - Home and office users are supported with free office software

- Linux is available in consumer products, such as Tivo recorders

Hardware required and supported

Modern computers have ample power for most computer applications. This is more true for Linux. At present there are versions of linux that will run on any platform from an IBM mainframe computer to a Palm Pilot. The smallest Linux installations can fit on a single floppy disk, and run on a diskless computer with a few MB of memory. On the other end of the spectrum, Linux will run on most high end computer systems. An average user would expect reasonable performance on a computer with an old Pentium 100 processor, 64MB of memory, and 2 GB of disk space. On newer computers the performance of the operating system is extremely fast. The list below gives some idea of the capabilities, but complete lists of supported hardware are available [5].

CPU - Intel family and clones, down to `386 processors

- Macintosh (Motorola)

- Others: Alpha, MIPS, Sparc, etc.

Memory - 16MB is a good minimum, 64MB is recommended

Disk - 200MB is a minimum, 2GB is recommended

Screen - Any size

Network - Any type

Others - Most PC hardware is supported - or will be soon

Applications and uses

By itself an operating system is somewhat useless, software applications are added to give desired functionality. Some of the common applications that a computer might be used for are listed below. Linux will support all of these applications, and more, with the right software [6].

Office - word processing, spreadsheets, etc.

Web and Internet Servers - host web sites

Server - databases and other institutional functions

Embedded - inside devices such as Tivo TV recorders

PDAs - an operating system for small handheld computers

Development - software authoring

Advantages and Disadvantages

A partial list of advantages and disadvantages is given below. The cost, stability and open nature of the system have been winning over a large number of corporate adopters. But, adoption has been slowed by people who don't understand the nature of free software or have a perception that it is difficult to use. In some cases there are also some software packages that are not available for Linux, and won't run under simulators [22] - the most notable of these applications are first person shooting games.

Advantages:

Free - paying for it is optional

Open - the source code is available and can be changed

Goodwill - developers and users are very helpful

Faster - it doesn't require newer hardware, extra memory and larger disks

Stable - it is very uncommon for Linux to crash (no blue screens)

Flexibility - more capabilities and features

Complete - all of the software is available and open - no `extra' software to buy

Security - very secure - unauthorized users can't change machine settings

Simplicity - point and click configuration

Disadvantages:

Compatibility - some programs will not run under simulators

Misunderstanding - some people believe `you get what you pay for'

Getting It

There are multiple distributions of Linux. While these all contain the Linux Kernel, they often include different pieces of software, and installation processes vary somewhat. The basic licensing agreement that Linux is distributed under requires that even if it is sold for a fee, it must be made available at no cost and it may also be copied freely by the user. As a result you can often download these distributions over the network at no cost [12][13]. The total download size can be up to 600MB. An alternative is to buy a distribution (the typical cost is $30) which includes a floppy disk, a CD-ROM and a brief manual. These can be found at any store that sells software. Sometimes the distribution will have a higher cost for `deluxe' versions - this more costly package often includes telephone support.

Distributions

In total there are hundreds of Linux distributions. Many of these are specialized for features such as embedded systems, foreign languages, internet servers and security. The list below is for user-friendly installation and usage. The most successful of these distributions is Redhat. Some distributions, such as Mandrake, are based on the Redhat distribution, but with enhancements.

Redhat - the original consumer friendly Linux [7]

Mandrake - a Redhat derivative [8]

Caldera - another well established distribution [9]

Debian - a release that focuses on stability [10]

SuSe - yet another distribution [11]

Installing

Each distribution of Linux will have a slightly different installation procedure, but they all follow the basic steps below. The total time to install Linux will between one to two hours. Users with a high level of knowledge can opt to do advanced setup, and new users will have the option of letting the system suggest installation options.

1. Turn off the computer.

2. Insert a provided floppy disk (you can also boot from a CD on newer computers)

3. Turn the computer on, it will start to load Linux

4. You will be asked some questions about the type of installation you want

5. Linux will format the disks, and start to load the software

6. While it is loading you will be able to set times, dates and passwords

7. You be asked to set up the graphics for the window manager

8. When the installation is done the computer will reboot, and you will be ready to use it

Some Terminology

The terms below are some of the keywords that are unique to Linux. These will appear during the installation, or during common usage of the system.

booting When a Linux computer starts it checks the hardware, and then starts

software. The process of booting takes less than a minute in most cases

kernel The core of the operating system that talks to all hardware and programs

shell A windows that allows you to type commands

permissions Control who can change what

GNU (Gnu's Not Unix) A group that develops free software comprising a large

portion of Linux

root This is the user name of the system administrator

File and directories

The directory and file structure of Linux is hierarchical, much like other popular operating systems. The main directory for the system is call root and is indicated with a single slash `/'. There are a number of subdirectories listed below that are used for storing system files, user files, temporary files and configuration files. A sample of the standard directories are shown below, and can be viewed with a file manager, or with keyboard commands. If other disks are used, such as a CDROM, or floppy disk, they are mounted under the root directory. (i.e., there are no `C', `A' or other drives, they are all under `/'.) (Note: the UNIX slash is `/', not the `\' used on DOS.)

A list of some of the more important directories follows with a brief description of each. Most users have their home directories under the '/home' directory. Most of the other directories are of interest to the system administrator.

/etc - device and software configuration files are kept here

/tmp - temporary files are created here

/home - user directories are kept here

/var - this is a place for log files, mail storage, etc.

/usr - software is installed under this directory

/dev - where devices are kept - they are accessed like files

/bin - some of the programs are kept in this directory

Every file and directory has a unique name which can be used to refer to it. Sometimes it is useful to be able to refer to groups of files without typing the name of each one. Wildcard allow file and directory names to be matched to patterns. The list below shows some of the wildcards commonly used.

* Any string

? Any Character

.. The directory above

. this directory

~ your home directory

Some examples of filenames with wildcards, and files they would match are shown below.

Ad* Advertise Advent.not Ad

Ad? Ad. Ade

Ad?.? Ade.d

??e.* ape.exe eee.thisisanother

../hi.* hi.there (in directory above)

~/*.there hi.there (in your home directory)

Filenames can contain numbers, letters and a few other symbols, but care should be used to avoid symbols that also have meaning to Linux, such as the asterisk '*'. File names that begin with a period '.' are system files that are normally hidden. For example, most users will have a file in their home directories called '.profile' or '.login'. These are used when a user logs.

Some of the standard Linux commands for files and directories are listed below. Most of the file and directory names can be used with wildcards.

cd newdir change directory to 'newdir'

pwd show present working directory

ls list the files in the current directory

ls -la list the files in the current directory in full form

ls files list files that match the 'files'

rm files removes the named 'files'

rm * removes all the files in the current directory (use with care)

rm /* removes all of the files in the computer (only do this if you are insane)

mkdir name make a directory 'name'

rmdir name remove a directory 'name'

mv from to move a file/directory 'from' an old name 'to' a new name

cp from to copy a file 'from' the an old name 'to' a new name

more file type out the contents of 'file' on page at a time

cat file type out the contents of 'file'

vi file a text editor for 'file' (some commands given below)

`dd' - cut a line (command mode)

'p' - paste a line below the current line (command mode)

`x' - delete a character (command mode)

`r' - replace a character (command mode)

`R' - replace a string (command mode -> edit mode)

`a' - append to a line (command mode -> edit mode)

`i' - insert a string (command mode -> edit mode)

`:w' - write to a file (command mode)

`:q' - quit from a file (command mode)

ESC - move from edit to command mode

cursor key - move the cursor

du check the disk usage of the current directory

du ~ check the disk usage of your home directory

df check total disk space available

sort this will sort the contents of a file

ln -s to from create a symbolic link 'from' a name 'to' a file

grep thing files search 'files' for the string 'thing'

compress file compress a 'file'

uncompress file uncompress a 'file'

User accounts and root

Linux follows very strict conventions for file and directory permissions. These require that each file and directory be given specific permissions for public reading, writing and execution. Each user is given their own account with a password, so that access to the system is controlled. Only the root user can access all files and directories on the system. Other users are limited to files they own, or files that have been marked public. Typically the root user is only used for administration, and normal users use non-root accounts. This generally keeps the system safe from careless damage, and security breaches. Each user has their own home directory, normally in the `/home' directory. The permissions for files and directories are set so that the user has complete control over that directory.

The permissions for files can be seen by doing a directory listing with 'ls -la'. This will show flags something like '-rwxrwxrwx jackh user' for a file that everybody can read 'r', write 'w' or execute 'x'. The leftmost 'rxw' is for the user 'jackh', the second 'rwx' is for the group 'user' and the rightmost 'rwx' is for everybody on the system. So if the permissions were '-rwxr--r--' everybody on the system can read the file, but only the owner can write and execute it.

For security reasons, write permissions for files are normally only allowed for the owner, but read permissions are normally given to all. Execute permissions are normally set for all users when the file can be executed, such as a program. Sometimes more than one user wants to have access to a file, but there are reasons to not permit permission to everybody. In this case a group can be created and given permission to use a file.

Commands that are oriented to users and permissions follow.

passwd user change the password for a user

chmod flags files change the permission 'flags' for 'files'

chown user files change the owner of 'files' to 'user'

finger user give information about a 'user'

who look at who is logged into your machine

last a list of the last users logged in

whoami give your current user name

su - name change to a different user

chgrp group files add a 'group' to a file

Most of the user information is stored in the '/etc' directory. For example, user account information is stored in the 'passwd' file. User passwords are stored in the 'shadow' file. Group information is stored in the 'groups' file. It is possible to add users to the system by editing these files, but there are commands that make it easier to update and maintain these files.

The 'passwd' command is used to change user passwords. In general passwords are the main line of defense against unwanted intruders. Most systems will do simple password checks when passwords are entered. In general, if a password can't be found in a dictionary or index of a book it will generally be safer.

Processes

At any one time there are multiple programs (processes) running on a Linux computer. When you run a program it becomes another process also running on the computer. Each process is given it's own unique process ID number (PID). Each process is given it's own private memory space, and allowed to run for a fraction of a second every second.

The list of commands below allow the processes in the computer to be seen. They also allow the general state of the machine to be determined.

ps -aux Print a list of processes running on the computer

kill -9 pid Kill a process with 'pid' running on the computer (uses the PID # from ps -ef)

passwd user Change the password of a 'user'

date print system date and time

who show who is logged into the machine

exit this will logout a user

fg bring background processes to the foreground

bg send a stopped process to the background

<CNTL>C hitting this key sequence will kill a running process

<CNTL>Z hitting this key sequence will stop a running process, but not kill it

command & any command followed by an '&' ampersand will be run in the background

Simple commands can be combined together with pipes to make more complicated functions. An example is 'ls | more'. By itself 'ls' will list all the files in a directory. 'more' is normally used to print out text files. But in this case the output of 'ls' is passed (piped) through 'more' so that it only prints one screen at a time. Multiple commands can be combined on a single command line by separating them with a colon ':'. For example the command 'ls ; ls ..' would list the contents of the current directory, then the parent directory.

Output from functions can be redirected to files instead of the screen. For example 'ls > temp' will take the normal output from the 'ls' function, and write it into a textfile called 'temp'. Input to functions can be directed into a program. For example 'sort < temp' will make the file 'temp' the input to the sort command.

Simple batch files can be created by putting a list of commands in a normal text file. The file can then be made executable using the command 'chmod 755 filename'. The program can then be run using './filename'.

Security

Security is not a significant problem for a computer that is not connected to a network, and passwords will protect it from `honest thieves'. When connected to a network there is potential for security problems. These problems become more serious when the computer is connected to the network 24 hours a day. General rules to keep a computer safe (this applies to non-Linux computers also) are:

keep user passwords safe - these can be the start of a security breach

protect the root password - loosing this throws the system wide open

shut down unneeded programs - network programs sometime have bugs that open doors

apply patches - software updates help close security holes

Shells

When one logs into a Linux system, you are actually running a program (shell) this is in some ways similar to DOS. In the standard shell you are given a prompt, where you type your command. If it is not a built-in command, it searches on the disk according to a user-specified search path, for an executable program of that name. Almost all commands are programs that are run in this manner. There are also executable shell scripts, similar to command files on DOS. Linux is limited to running a program of a size equal to the sum of its memory, and swap space. As the system is multi-tasking, any program (or part thereof) that is not currently being run when extra memory is required, is swapped (moved) out to the disk, until it is ready to run again.

In shells there are environment variables set. Some of the commands that can be used to view these are shown below. They can be set by editing the appropriate text files.

alias prints a list of command aliases

printenv prints a list of the environment variables

set prints a list of the environment variables

X-Windows

The GUI in Linux is actually two programs working together. The basic program is called X windows, and it provides basic connection to the screen, mouse, keyboard and sound card. The look-and-feel of the GUI is provided by the window manager. One simple window manager is called `fvwm' and it can behave like Windows 95/98. Newer window managers include Gnome and KDE. While these both provide similar capabilities and features, most users develop personal preferences for a single window manager.

Configuring

Devices and settings can be configured under X-windows using graphical tools. Settings can also be configured with text files, but this is not necessary. Examples of settings that the user or root might want to change are:

Modem properties for internet connection

Network card properties for connection to a LAN

Printer type and location

Customize the windows settings and behavior

Sound card settings and sounds for Window events

Desktop Tools

Most users focus less on the Operating System, and more on the programs that it will run. The task list below includes many of the applications that would be desired by the average user. Most of the listed applications are free, with the exception of the games. Many of these packages are a standard part of Linux distributions.

· Office Software - these include spreadsheets, word processors, presentation software, drawing

tools, database tools, 3D graphics tools

Star Office [14]

KOffice [15]

· File and Internet Browsers

Netscape - allows browsing of the internet [16]

Files - there are many file viewers that ease directory browsing

Eazel - allows active directory browsing [17]

· Administration and Utilities

Apache - the most popular web server program [18]

Postgres and MySQL - Database programs [19] [20]

Replace a microsoft networking server [21]

DOS/Windows Simulator VMWare [22]

· Entertainment

Audio and video

Tools (GIMP - similar to photoshop)

Games (Quake, Doom, SimCity)