Implementing Robotic Automation

Getting Started

Getting started in evaluating whether robotics is the right thing for a firm begins with motivation, typically through the means of identifying pain in the manufacturing process. Pain comes in many forms such as waste, scrap, ergonomics, high cost per piece as compared to competitors, and finding reliable operators. Based on many years of selling automation, the following process seems to work well:

Step 1 : The firm has identified pain somewhere in the process

The firm has motivation and management support to pursue robotics as solving a problem

Return on investment, hurdle rate, and/or payback period requirements for the project are identified The firm has a reliable process

Step 2: Defining the process:

If this is the first robot project, identify a low-risk process and product that will be an immediate win upon installation

Identify product(s) to be automated

Identify project design criteria (the constants). In other words, what are the expectations and requirements for the project parameters

Step 3: Develop concept:

(This exercise serves as a litmus test for the merits of the project)

Establish a rough order of magnitude of costing for all equipment and services

Define a preliminary automation solution Determine responsibilities within the firm as well as out

Identify the unknowns

Identify where the system will be located Identify resources to operate the system side integrators, contractors, etc.

Step 4: Decide whether the project makes sense economically and technically:

Fgr. 1 shows a typical flow chart for implementing robot projects. This section deals with the audit/discovery aspect of such a project.

===

Identifying risk/unknowns

Inquiry

Confirming the brutal facts

Cost lo manage risk financial/technical

Process requirements

How Expectations operations function

Resources required

Commercial definitions

Equipment function defined

Identify unknowns and risk

In-house expertise

What are the gaps

How would you process manually

What is in within your control

Audit/Discovery:

Internal and external drivers

Information gathering critical factors

Judgment Criteria

Defining constants and variables

Upstream and downstream

Existing or new process

* Goal setting /Validation

Where it’s being done elsewhere

Feasibility study

Manage tolerance stack-up

Validating process variables

Risk/reward checks

When a project is not a good project

Area of Focus -------

- - . Design/Build

Commissioning

Purchasing/Timeline/ Project Management

Fgr. 1 Best Practices- Life Cycle for Implementing Robotic Projects

===

Robotic Implementation is Always a Team Effort

There is no question that a team is necessary to implement a robot project, because many disciplines are required to add value in the stages of planning and implementing such a program. To be most effective, the plan must be developed by a team, as with many programs, For example, an IS0 quality management program. The personal profiles of a team that would be ideally suited for developing a robotics plan are as follows:

Job Role

[ Production Project Manager Maintenance Engineering/Design Facilities

Finance Programming Safety]

Contribution to the plan for industrial robotics

[ Primary point of contact and communication with suppliers and "Champion" internal customers such as operations people. Manages schedule, develops scope of work. Personnel are typically manufacturing engineers, program managers, project managers, quality managers, and sometimes technicians.

Manage the system on a daily basis, track system uptime, manage process consumables such as pallets, slip sheets, welding wire, and grinding media.

Manage daily system problems. Personnel are typically operators who were involved in the process prior to the installation of the robot, or technicians.

Installation of the system, and maintenance over the life of the system. Personnel are typically maintenance technicians Design work-holding, fixturing, and product design to enable the process to be "robot and operator friendly." This role is typically associated with process development. Personnel are typically from the engineering group (i.e. Industrial Engineering)

Manage logistics for utilities, installation, and floor requirements. Facilities may also be responsible for building components such as electrical panels or fabricated components.

Validate the ROI and put together the justification package. Typically other personnel also assume this role.

This role consists of programming for system controls and robots. Personnel are typically engineering, operator, technician, and maintenance people.

This role manages the safety compliance of the equipment brought into the facility. There maybe a safety director, or safety may be managed by others. ]

Teams will vary based on the scope of work and organization size. The roles and responsibilities are the same across any organization implementing robotics. Other roles and responsibilities of the team include supplier selection for any outside-sourced components, process development, especially if the process is new, and criteria for justification including quality, reduced direct labor costs, productivity increases, safety, etc.

Other members of the team are often outside suppliers or consultants. Especially for first-time robot users, system integrators are often contracted to provide a "turn-key" automation solution to accomplish a task(s) within a manufacturing process. System integrators are organizations that have the capability of managing the qualifying, justification, quoting, project management, installation, and commissioning of a robotic system. The firm will also rely on the system integrator to train their employees in the use of the system, as well as adapting additional products into the system. The system integrator organization is set up to include all the disciplines mentioned above. The relationship between the manufacturer and the system integrator will vary for every project, and the relationship is based on defining which organization is responsible for various attributes of the project. Trade-offs always need to be considered when dealing with a system integrator in terms of how much risk the manufacturer is willing to manage, the internal versus external costs of designing and building system components, delivery needs, internal capabilities for services that would otherwise be pro- vided by the integrator, and availability of resources within the firm.

There is no wrong or right decision.

Consultants sometimes provide equipment such as hardware, but often are contracted to provide design or project specifications that will facilitate a request for quote from a system integrator, or to write the execution plan if the system is to be managed internally within the firm.

===

Process machinery (lathes) (machine tools)

Part turnover stand to allow robot to re-grip part between machining and deburring

Robot fixed on a fabricated steel riser abbot uses a two part gripper

Deburring station utilized after machining

Fencing (barrier)

Inbound powered conveyor for raw parts (8 part queue)

Operator system control

Outbound powered conveyor for finished machined parts (8 part queue)

Fgr. 2 Anatomy of a Robotic Machine Tending System

===

Welding wire feeder powered by welding power supply Robot/-Fabricated Steel; Welding Torch (end of arm tool)

Positioner controller by robot x 2 Work-piece fixturing installed here safety (positioner serves as the inbound and outbound) - work-piece loaded manually to fixture \ Light Curtain for

Fgr. 2 Anatomy of a Robotic Welding System

===

Existing conveyor that feeds the robot system The gripper picks slip sheets, pallets, and product Fgr. 3 Anatomy of a Robotic Palletizing System

====

Powered inbound conveyor; Powered outbound conveyor; Robot 1 Robot gripper (end of arm tool)

Part qualifying tool on stand -System control components /Multi-purpose belt stand with various grinding belts; Fencing

Fgr. 4 Anatomy of a Robotic Material Removal System

===

Robot controller transport unit Press Brake Part reference stand-over stand / System Control Inbound floor locators for pallets -- Sliding gate Fence (Barrier) floor locators for pallets with formed parts

Fgr. 5 Anatomy of a Robotic Press Tending System

===

Ingredients of the Robotic System

Figures 2 through 6 illustrate a typical robotic system for machine load/unload, arc welding, palletizing, material removal, and press tending. Although the applications are different, the ingredients that are contained in the system are similar. There is always the consideration of transferring raw material into, and finished product out of, the system. For each application, the means is different, however the ingredient is still required or the system is incomplete. The point is that, regardless of the application, the planning and auditing of a process is the same because different robotic systems require the same ingredients, which are illustrated in Figures 2 through 6.

Regardless of the application or industry, the integrated components of a robotic cell, as well as the planning process, are virtually identical, although each application has unique process-specific characteristics. For instance, applications such as robotic press tending differ from robotic arc welding. Essentially then, the planning process for one robotic system can serve as a model for future robot systems, regardless of the application. It’s worthwhile to understand the typical ingredients for planning and purchasing a robotic system. The recipe for automation follows the similar rules regardless of the application and the integrator source, whether internal or external. Various system elements and their contributions to the system are shown in the list below. Not every system will contain all of these components, so some items may not be applicable. Nonetheless, careful consideration is needed to develop a robotic system shopping list of the items that are illustrated in figures 7 through 15, and the items fall into the following categories:

===

Ingredient (Category Component); Robot

A sir axis rail mounted configuration Six axis robot size and reach range example Robot reach 1 meter radius -- 3 meter radius Robot payload 11 lbs. (5Kg) Exceeding hundreds of lbs

Fgr. 7 Contrasting Robot Size, Reach and Configuration

===

Role/Responsibility

The robot is the manipulator that performs the sequence of operations within the process. Throughput requirements establish the number of robots needed. The process establishes the robot criteria in terms of reach requirements, number of robot axes, weight-carrying capacity, and robot configuration to support such specialized applications as vision, auxiliary axis, interface to machinery or process devices, and other needs.

Inbound

Welding headstock/tailstock positioner ; Pallets on Floor containing materials ; Tray/Drawer system ; Various Conveyor styles

Fgr. 8 Inbound and Outbound Material Transfer Devices

Role/Responsibility

Inbound describes the hardware that enables raw material to be transferred into the system. Examples include conveyors, turntables, shuttles, AGVs, and fixed stations in the form of a table, or poor location.

Outbound

Role/Responsibility

The hardware that enables the finished material to be transferred from inside to outside the system. The hardware is similar to that for the inbound

End of Ann Tool (EOAT)

Fgr. 9 Robotics end of arm tool configuration.

Role/Responsibility

The gripper or "end of arm tool" is the functional tool used by the robot to perform the process task. Tools are part- and process-specific. Some processes such as welding or cutting, can often use a common tool. Tools used for material handling or material removal often are varied from process to process.

Tool Changers

EOAT Changer Plates with Storage Rack

Fgr. 10 Robotic Tool Changers

Role/Responsibility

Tool-changers are quick-disconnect (male/female) devices that allow the robot to change tools automatically. Changing tools allows for different processes to be performed by the robot as part of the robot program.

Auxiliary Axis System Fgr. 1 1 Auxiliary Robot Axis Headstock/tailstock positioner

Robot transport unit 2 axis tilt 8. rotate positioner Robotic Servo Boom

Role/Responsibility

Auxiliary Axes are additional axis of motion controlled by the robot control system. The aux axes are programmed through the robot control system. Aux axis can control the motion of the part or the robot to the part through the process. Many additional axes can be added to the robot control for the purpose of providing a greater level of flexibility in the robot 5 process.

Controls

Fgr. 12 Robotic System Controls Role/Responsibility

The control system serves as the interface for the human operator to manage and monitor operation of the system. For example, some users require what is called "Overall Equipment Effectiveness ". This function requires a system control to gather system data and report the data related to downtime, faults, uptime, top 5 reasons for faults, and other production data that allows the user to track the efficiency of the system.

The control system is also used for information transfer and communications between all the devices in the system, as well as controlling the robot. The control system can be PLC or PC based, or can be within the robot control. Often the interface to the control system is referred to as the Human/Machine Interface (HMO or Graphical User Interface (GUI). Selecting parts to be run in production, collecting and archiving system data, monitoring system parameters, and transferring/receiving data within a larger- scale network are additional functional requirements of some system control schemes Larger scale control systems utilize Supervisory Control and Data Acquisition (SCADA) systems. These systems are responsible for information management at the system and remote from the system. These systems also can interact with Oracle, and SAP databases for downloading production schedules for the robotic system as one example.

Fabricated components

Fabricated platform containing the automation for machine tending

Fabricated turntable for robotic glass handling

Robot risers

Welding positioner

Fgr. 13 Fabricated Peripheral Components

Role/Responsibility

Components that are necessary to the robot system, usually in a supporting role for the process Examples are robot risers, positioners, turnover stands to enable the robot to re-grip a workpiece with a specific orientation for machine loading, stands, racks, reference tables, storage equipment, and other items.

Process equipment Press Brake Welding Equipment Machine Tool Lasers

Fgr. 14 Process Equipment

Role/Responsibility

Process equipment is a very broad group. Within the process there is machinery that does something to a part, or provides the robot with the capability to perform a process such as the welding equipment for welding. Other examples include tending to a machine tool cycle, forming on a press brake, or operating a laser for cutting or welding

Safety Guarding

Role/Responsibility

Consists of wire mesh fencing, light curtains, safety mats, limit switches on the robot arm to define robot motion, sensors and switches on conveyors and access gates. Typically any device that limits the ability of the robot and the human operator to share the same space at a given time. Safety can also include custom flooring to prevent slipping in a hazardous environment, or a smoke exhaust system to remove welding particulates.

Miscellaneous items

Role/Responsibility

An even broader classification. Examples include inspection devices, gages, washers, dip-tanks, pallet dispensers, welding torch reamers, stretch wrappers, pin stampers, part markers, bar code readers, or any piece of equipment that is integrated within the sequence of operations for the system.

Devices to locate the work-piece Role/Responsibility

Robots are blind. Requirements to locate the work-piece from a casting, tacked welded assembly, steel blank, or bag of seed corn, is concerned with managing the tolerance for locating through the sequence of process operations. Examples include welding fixtures, powered conveyors with V-block locators to hold a shaft, or pushing devices at the end of a powered roller conveyor that moves the bag of seed corn to a specific location at every cycle.

Protection Tools

Fgr. 15 Robot Protection Tools

Role/Responsibility

Typically designed to protect the robot from a harsh environment.

These items include robot suits for a wash-down environment where oxidizers or other liquids are used to sterilize and clean the robot.

Additionally, NEMA 4 enclosures, severe dust and liquid hazards for equipment, and cooling or heating robot suits are required.

Installation requirements

Role/Responsibility

The services required to install the electrical and mechanical components of the robotic system from utilities to floor requirements.

Internal or external integration services

Role/Responsibility

The man hours required to design, build, integrate, de-bug, and manage projects, and validate, document, and install the components of the system.

Training

Role/Responsibility

Training is imperative. An educated user is absolutely one of the most important items on this list of ingredients. Training in managing system faults, and general error recovery is critical in maximizing system uptime.

Programming

Role/Responsibility

Every system requires programming of the robot through the use of a teaching pendant, connected to the robot control or off-line within a PC environment.

Robot motion is an aspect of programming in addition to the system controls, to ensure that the sequence of operations is executed seamlessly.

Vision and Sensors

Role/Responsibility

Vision has simplified robot cells and at the same time increased flexibility, especially for small batch runs. Sensors and vision can detect presence or lack of presence of physical features, and can determine part location and orientation. Vision and sensors are also used to validate in-process quality, in optical character reading, and in part identification

The shopping list is a tool to begin organizing a scope of work for the system. The robot itself is just one component of the list.

Defining the robot is normally the simplest task of the project. The system risk and primary cost are driven by all the peripheral ingredients, especially the criteria for the transfer of raw material in and out of the system. The saying " you can do anything with automation" is a true statement, it just takes money. Developing a scope of work that defines the ingredients in a robotic system, as well as the functional specification, makes all the difference in reducing the system risk and cost. The costs of robot systems are a function of the part requirements, process, level of flexibility, and the services responsibility. The system hardware is generally the major part of the system cost, as compared with the servicing hours associated with the engineering, management, and integration of the system. It’s easy to see why automation requires a team effort, and why there is a large market for system integrators that manage the details. The devil is in the details and the more details can be defined early in the plan, the lower will be the integration costs, and the more successful the system.

Judging a Good Project from a Bad Project

For every application type, sources are available to serve as tools for defining the needs and wants of the project. At the start, the most important first step is to identify the need. In other words, the source of pain in the manufacturing value stream needs to be identified.

The conclusion may be obvious, but there is an important exercise that first needs to be performed.

Before robotics can successfully be implemented for a particular process, the engineer needs to understand the manufacturing sequence of events. This exercise is important because the sequence must be defined before the process can be integrated with robotic automation.

Furthermore, the relationship between one sequence and another within the value stream, from the receipt of raw material until the product is packaged on a pallet ready to ship, also needs to be defined. Automating an inefficient step, or the wrong step in the value stream, can backfire quickly, and the gains the user assumed would be received won’t be realized. In other words, never automate a bad or undefined process.

A common approach to planning automation is to examine how the process needs to operate in a manual operator state, because most engineers already understand how to manage the manufacturing process with direct labor. Whether the manufacturing process or value stream is already in place, or is being redefined because the process is new, the exercise of defining the process with manual labor allows the engineer to determine many factors. These factors include throughput requirements, labor requirements at each step, cycle time of each step, and the overall sequence of events. At every step it’s necessary to examine how waste can be removed and reliability added in.

Defining the process using direct labor establishes a reference to which automation results can be compared later. The following criteria are applicable for most processes, both those in place, and, if the process is going to be new, those to be included in a new order.

Some important questions regarding a hit list of process information are as follows:

What is the throughput required in units per day or shift to meet demand for each of the individual parts to be produced? How many hours need to be planned on a daily basis to meet this demand? What are the sequential steps in manufacturing the component? Process sheets outlining the operating steps are very useful here

How much labor is needed at each step to achieve the overall throughput? What changeover requirements are necessary for the operator to switch from production of one product to another, and how much time is associated with the changeovers? operator to produce a compliant or quality part? the frequency of inspection? What decisions need to be made at each step in order for the

What kind of inspection is required in the process and what is

What are the characteristics of a "good" part at each step?

Are there any safety, ergonomic, or environmental concerns to the operator?

Are there redundant operations in the process where the operator task is the same, over and over, for every cycle? What is the scrap rate, and where do quality problems lurk.

Understand the root cause of any quality problem. Is the product within tolerance? Are there any current floor space, utilities, or ceiling space constraints for where the process is located today, or where it will be located later?

What is the means of material flows from operation to operation?

Are part prints and models available?

Other benchmarks that need to be identified, especially for existing processes, are related to achieving the intended throughput rate.

End-users often say "I am getting 45 parts per hour". The problem is that the available time on an hourly basis indicates that the throughput should really be at 65 parts per hour. Overtime rates and the number of shifts also should be taken into account because robotics will impact efficiency, resulting in a higher throughput using the same quantity of assets. The reverse could also be important, meaning that even though the robot system is capable of producing 65 parts per hour, the customer demand is for 45 parts per hour and there is no benefit to building up inventory.

The heart of the matter is: "is this a good or bad robot project"? While going through the exercises above, the engineer will learn a great deal about the process, whether it’s new or existing. It’s best to answer the questions directly to users. However, it's usually obvious if the robot project is not a good project. There is a quick mortality rate for bad projects because right away there will be indicators, often with the shop floor operator struggling, and there is no sense in automating a bad manual process. Examples are lack of chip control in a machining operation that causes chips to build up at the work-holding locations and prevents the robot from loading the work-piece.

Another example is weldments with a gap condition that exceeds the welding wire diameter several times over, or welds that simply cannot be accessed. The other immediate feedback is when the economics and/or justification shoot down the project. Also, there is sometimes a better way to skin the cat than to use a robot. Not every process should be automated, and for some projects the economics make no sense versus alternative forms of automation, or the cost is too high to manage all the sins of the process. Some examples of applications that can be bad candidates to automate are as follows: The process, by design, prevents parts from meeting print tolerances Operator intervention is very high relative to decision making and dexterity. Examples are special inspection requirements that a conventional gage or sensor cannot measure, or when the inspection is more visual than empirical, or there is a need to adapt to varying welding part conditions, that require the operator to compensate to make an acceptable part at each cycle Changeover requirements are too drastic, so that a significant amount of the daily hours available are used up in changing or adjusting hardware. There certainly are jobs like this, but not generally. The problem is that the robot efficiency can never be exploited because the process does not allow enough run-time.

Another form of "dedicated" automation is a better fit because of too much variation from part to part in each cycle, or the same redundant process is used in a high-volume setting. Examples are a linear pallet pool (LPP) system for a flexible machining system that is machining a batch of one production flow. At each cycle, a new part with unique geometry is introduced into the system. Again, there are exceptions where robots will replace the LPP as the primary device servicing the machine tools. Another example is a high-volume lathe-welding process where end caps are welded to each end of a pressure vessel. Typically a dedicated form of automation will be used to accomplish the "common" task with a fixed welding torch and the work-piece turning 360 degrees under the fixed torch.

Floor space or other external constraints make automation impossible. Often in food-focused industries, For example in bakeries, the infrastructure of process equipment and space was never planned for automation. Older facilities are especially land-locked in their ability to support the space for a robotic system.

Automating a process with significant downtime caused by poor running equipment, or a poor process.

Automating a process where the operators or production culture are not sold on automation. Attitudes are among the most critical characteristics governing whether a project is going to be good for automation.

Many more examples of good and bad projects are discussed in this text.

The next section describes the importance of gathering details when the project parameters and solutions pass the litmus test in terms of economics and technical feasibility.

Gathering the Right Details

The detail that is gathered to justify or to initiate a robotic project really is the same across all industries and applications. There are process-specific and applications-specific criteria, but the methodology is the same.

The first thing that should be gathered and the item that drives the entire process, solution, and cost, is the parts/products that are being manufactured, and specifically the process steps considered for automation. This first thing may take the form of prints, albeit 2D, but availability of 3D solid models makes a big difference in the type of analysis that can be developed. The old GIGO saying that garbage in is garbage out is absolutely true when it comes to gathering the detail. As a minimum, a start should always be made with the prints.

The applicable checklist for any application is the list detailed in the previous section. The following pages describe a more detailed list for various application types:

===

Criteria for Arc Welding

Example of a Weld Procedure Table

Weld Number XI

Example of a Weld Procedure Table

===

Weld procedures (gas type, filler metal, weld travel speed, bead size, joint type, number of weld passes per joint type)

Preheat, post-heat, or inter-pass temperature requirements Work-piece positioning to enable each weld to be welded in the flat or horizontal position Quantity of welds for each assembly or sub-assembly, and length of weldments Product weights Product base material Pre-tacked parts or loose parts. Can loose pieces be locked Welding sequence for distortion management Tolerance of the weld joint (dimensionally, and as a gap condition of the weldment)

Managing fume collection

Definition of a "good" weldment (i.e. penetration, size, Definition of inspection criteria if applicable

Floor layout constraints

How the system interacts with the operator for part and tabbed to simplify fixturing appearance ) changeover and part setup prior to robotic welding (i.e. tacking, heat treating, prepping work-piece)

Monitoring of process parameters and deciding which criteria is important to monitor. What kind of feedback is being measured, and what actions are to be taken when a parameter falls out of specification? For example, the variations may be in a flow switch controlling water flow to a water-cooled torch or a flow meter controlling shielding gas flow to the weld area.

How to balance operator load/unload time of finished and pre-welded assemblies while welding continues Safety precautions to protect the operator without interrupting the robot weld cycle during the load/unload cycle The critical component for the robotic welding cell is the fixture holding the work-piece. There are two primary fixture styles; holding fixtures that hold a tacked assembly in place, or a tacking fix- ture where the pieces are loosely held and the robot tacks and welds.

Fixture design affects robot access, torch access, changeover from one part style to another, weld cycle time, part tolerances, and cost.

Access to the weld areas of the work-piece, and part dimensional tolerances, are critical to reliable welding. These requirements are the reason why manual fixtures often don’t work for robotic applications. Fixturing is the key to a successful welding program.

===

Fgr. 16 Weld Fixturing Examples

Multiple work-pieces on two sides of a fixture allow a balanced throughput between operator load/unload time and welding time. In this fixture the operator loads both sides of the fixture.

This positioner will be integrated to a rotating robotic headstock/tailstock positioner. Parts are not tacked when located in the fixture

This fixture is an example of a holding fixture that secures the part with open access for the robot to weld. This fixture will be installed within a robotic headstock/tailstock positioner

===

The fixture designer must be aware of how the process is defined and where the robot and weld torch need to be positioned to achieve an acceptable weld. Whether the fixture is automatic or manual is another consideration that needs to be addressed. How the operator interacts with the fixture needs also to be understood. In other words, how automated is the clamping function of the fixture.

Designing for quick changeover of fixtures also is necessary if multiple parts styles are to be welded in the system. Fixtures must be designed to resist weld spatter. Fixtures may also have to be designed to confirm part presence, to validate that the part is secured correctly, and that it’s located within the tolerances needed for welding.

Criteria for Tending Machines

Part weights

How the machine tool work-holding is holding the work

Type of work-holding for each process

Dimensional tolerances of component and tolerance for loading machine fixture(s) at each step Required orientation of the work-piece when the machine is loaded for each operation

Base material for each component to be machined

List of sensitive part surfaces that must not be touched by robot grippers

Can parts be allowed to touch in the queue at the inbound or outbound sides of the conveyor

How large a part queue is needed at the inbound side

Cutting times for each process step

How are chips managed in the fixtures and discharged from

Machine tool type and quantity at each process step

Do the machines have shuttles or pallet changers

Machine tool quantity by type for each cutting operation

Machine tool control brand and model

Where do programs reside for the machine tool

Signals for hand-shaking between robot control systems and Gauging requirements or other inspection tasks

Washing or deburring requirements

System controls for feedback and monitoring of process piece for each operation the machine tool machine tools parameters, and how operator interacts with the system for changeover

Air blow-off

Floor layout

Requirements for access to machines without interrupting robot cycle

How parts are presented to system (drawer system, conveyor, dunnage, shuttles, bin, pallets, etc..)

How parts are transferred out of the system. If they are palletized, what is the design of the pallets, and what is the palletizing pattern.

Managing parts that fail in the machining process where do they go

Keeping the floor clean of coolant spillage

Buffer requirements in the event of uneven machining cycle times between processes

Time study of machining process

Can chip breakage be incorporated in the tooling

Does the work-piece require a pusher or other mechanism for reliable loading

Is a washing operation required after machining

What is the tool life, and the frequency of tool adjustment or changing

Will an air blow step suffice to separate the part from chips or is a washing operation necessary

How do offsets get updated in the machine tool

An example of a machining time study is shown on next page:

The key to machine tending, and where the bulk of the cost lies, is associated with the transfer of product into and out of the work-cell. Machine tending is ideal for running lights-out, depending on the design criteria of the inbound and outbound systems. A good approach is to design the raw material inbound queue to enable the system to run a reasonable amount of product without an operator presenting new material and removing finished material. Typical queues are planned around 30 60 minutes. For long cutting cycles, or parts that can be picked directly from a bulk container as an example, production can run for significantly longer unmanned periods. See Fgr. 1 8 for an illustration of bulk part presentation.

Criteria for Press Tending:

Part weight

Thickness of blanks

Base material

Painted surfaces

Oily, sticky, surfaces of raw material

Press type ( tonnage, bed length, control system, gage system ) How product is transferred into and out of the work-cell

Are formed parts palletized, and what types of pattern or nest are used

System control requirements for production monitoring, part selection, and part changeover

Defining brake tooling for the die and punch

How each part type is processed through the bend sequence, indicating the bend

Bending tolerance and inspection requirements if applicable

Changeover requirements for brake tooling

How will tooling be set up in the brake ( i.e. progressive tooling )

If the base material is stainless, is there a plastics covering on one side

Floor layout

How is tooling changeover managed

Unlike machine tending, where the emphasis of cost is in the material transfer into and out of the work-cell, the inbound queue for press tending is much simpler because the raw material is flat blanks that are stacked on top of each other. Press tending is ideal for a small-batch process where the system is designed to be automated for many part styles. The key and emphasis on cost are related to the grippers that are required to handle all the potential part geometries through the forming processes. Additionally, in brake forming, how will the tooling be con figured for small batch runs. If changeover is minimal, then a high uptime can be assumed. It may also be possible to use the robot to change brake tooling to reduce changeover time.

Palletizing/De-Palletizing Data

Pallet types and sizes

Slip sheet types and sizes

Description of each product ( size dimensions, weights ). Are they boxes, buckets, containers, etc.

Unit load description (number of layers of product, are layers of product flipped every layer, product per layer (pattern)) If bags, what material (polythene, paper, etc.)? Wash-down considerations, or special environmental requirements for cleaning or hazardous specifications

Floor layout If cases are used, what kind of cases. Do cases need special handling to avoid damage. Do labels need to be presented in a certain way when the case is placed onto the pallet How pallets are presented to the system How are slip sheets presented? Mixed-unit load requirements Are stretch wrappers or other wrapping methods required? How is raw material presented to system (i.e. conveyor)? What is the control system for production monitoring, and Is integration to a higher plant level control required Managing buffer in a system when a full pallet is transferred How are full unit loads transferred out of the system If de-palletizing, what is the tolerance for product location as What tolerances are required for the inbound raw material to Product inspection prior to reaching robot cell, and arrange product changeover out of the cell it rests on the loading unit build a uniform unit loadments for dealing with non-compliant product. (i.e. checking bag weight as bags are conveyed to robot)

Definition of a "good" unit load Flap management requirements (i.e. flap detection)

Use of lock'n'pop or other means to prevent cases shifting during transport Palletizing and de-palletizing systems need to be con figured primarily around transfer of raw materials in, and finished unit loads out, or vice versa. Palletizing can be cellular where the cell is fairly simple, as shown in Fgr. 3. Robotic palletizing can also become very complex, especially in a high-volume palletizing facility with multiple product lines and/or product, converging on a centralized palletizing area. The complexity involves a high level of material transfer equipment such as conveyors and high level system controls to manage the additional products and maintain communication between devices.

Criteria for Material Removal

Material removal is a very broad term but for the purpose of this text it’s considered to include deburring, grinding, de-flashing.

Buffing, and polishing are related processes.

Defining a good part or surface finish/condition for every region on the work-piece to be processed Does a constant force need to be applied to the surface to achieve surface finish requirements Defining the part tolerance where removal of surface material is required. For example if the work-piece is a casting then the tolerance of the surface to be removed will vary Identifying where material is to be removed from the workpiece

Defining the range of product by size, weight, and base material

The physical size range of burrs, gates, flashing, and parting line that is to be removed

How product is to be transferred into and out of the work-cell

Tolerance for how much material is required to be removed

Defining the media that is currently used manually (type, relative to the base material style, and grit)

Defining the tool life and frequency of changing tools

Feed rate in removing material

Base material type

Material removal is defined in two ways. Fgr. 19 illustrates an example. The media or also called tool (e.g. Snag Grinder) is taken to the work-piece or the work-piece is taken to the media or tool. In either procedure, compliance is also a major factor in providing a reliable robotic process. Compliance is defined as maintaining a constant force during the material removal process, taking into account such factors as the variable surface tolerances of the work-piece, gravity, the position of the tool, and media wear.

Material transfer into and out of the work-cell is also a large component to the system concept, just as with the other application types.

===

Tool (media) to part

Part to tool (media)

Note: Smaller parts are often presented to the tool (media). Larger parts are usually fixtured and the robot takes the tool (media) to the part

Fgr. 1 9 Material Removal Examples, Tool to Part or Part to Tool

===

The need for robotic material removal is more important than any other process because of safety, quality, direct labor cost, scrap reduction, and difficulty in finding and training operators. Material removal is a brutal job to perform manually. Usually new employees get material removal duty first because it’s a brutal job. At the same time, material removal poses the greater risk compared with other robotic applications. Risks are associated with defining a "good part" and the trial and error that goes into developing a reliable process, given varying part tolerances. Most material removal projects require a validation phase whereas for other applications the validation phase is minimal or not required.

A good start in assessing how robotic material removal can be accomplished is to first examine the tools currently being used by the manual operator and process. Fgr. 20 illustrates a before and after example where the robots tools were designed to mimic what the operator used.

Before automation; After automation

Fgr. 20 Applying the Tools Used Manually for Robotic Automation

Constants and Variables of a Robot System

Constants are defined here as what is known about the process as well as what are the rules. The data that is gathered during the audit phase of the project is termed design criteria. Design criteria are the known parameters of the process. Examples are noted below.

Understanding the rules about the process requirements is what drives objectives, goals, expectations, and benchmarks. The variables are associated with how the solution will be conceived, designed, and built. At some point in implementing a robotic system, the system has to be validated in terms of function and performance to ensure that the design of the system solution achieves the goals within the rules set early on in the project audit phase.

Providing answers to the application question lists in the examples establishes very quickly how well the process is defined and what it’s that you know and don't know. Most expectations fail to be achieved because the constants are not accurately defined up front.

During the audit phase it seems that many engineers are rushed into a time frame to implement a system where details are overlooked, ignored, or simply guessed. An example of establishing constants and variables during the audit phase of a project is as follows:

Examples of Constants (Design Criteria):

Defining the generic manufacturing process by sequence of Product information (including prints, solid models)

Description of how product needs are to be transferred into Description of special requirements in locating parts for the Description of what happens, and the requirements of each Description of manual intervention at each operation Process sheets for each operation operations the system robot process operation

Examples of Variables (Automation Solution):

Robot process achievement rate

Robot access through all the robot motion in a program, Robot gripping in terms of feasibility between how the product needs to be gripped and what can be designed avoiding obstacles and having sufficient reach

Achieving material quality requirements (weld profile, surface finish as examples)

Functionality of peripherals. For instance, will vision work in the shop floor environment, or will the fanning magnet actually separate the steel blanks so that the robot doesn't pick two blanks instead of one at the beginning of the process

Defining the project design criteria using the example project shopping lists is a good habit. One of the by-products of the defined design criteria is a set of variables that make up the automated solution. The variables (unknowns) are sorted and ranked by various levels of risk. Managing the unknowns is what separates the good and the better system integrators as well as the users. Unfortunately, in some instances, the unknowns are not even known until the system is built and in a position to be run. Having a plan to manage risk ultimately improves cost, delivery, and reliability of the robotic system.