MANUFACTURING PROCESS PLANNING AND DESIGN:PROCESS PLANNING

PROCESS PLANNING

As noted above, process planning is a function that prepares the detailed work instructions for a part. The input to process planning is the design model and the outputs include processes, machines, tools, fixtures, and process sequence. In this section, the process planning steps are discussed. Please note that these steps are not strictly sequential. In general, they follow the order in which they are intro- duced. However, the planning process is an iterative process. For example, geometry analysis is the first step. Without knowing the geometry of the part, one cannot begin the planning process. However, when one selects processes or tools, geometric reasoning is needed to refine the understanding of the geometry. Another example is the iterative nature of setup planning and process selection. The result of one affects the other, and vice versa.

Geometry Analysis

The first step in process planning is geometry analysis. Because the selection of manufacturing processes is geometry related, the machining geometries on the part, called manufacturing features, need to be extracted. An experienced process planner can quickly and correctly identify all the pertinent manufacturing features on the part and relate them to the manufacturing processes. For example, the manufacturing features of holes, slots, steps, grooves, chamfers, and pockets are related to drilling, boring, reaming, and milling processes. The process planner also needs to note the access directions for each feature. Also called approach directions, these are the unobscured directions in which the feature can be approached by a tool. When features are related to other features, such as containment, intersection, and related in a pattern, these relationships must be captured. Feature relations are critical in deciding operation (process) sequence. Figure 4 shows a few features and their approach directions. The pocket at the center and steps around the center protrusion are ap- proachable from the top. The hole may be approached from the top or from the bottom.

In computer-aided process planning, the geometry analysis (or geometric reasoning) is done by computer algorithms. The design model in the form of a solid model is analyzed. Based on the local geometry and topology, regions are extracted as features. To be significant to manufacturing, these features must be manufacturing features (Figure 5). Again, manufacturing features are geometric entities that can be created by a single manufacturing process or tool. Like manual geometry analysis, geometric reasoning must find feature access directions (approach directions) and feature relations. Due to the vague definitions of features, the large number of features, and the complexity in feature matching (matching a feature template with the geometric entities on the solid model), geometric reasoning is a very difficult problem to solve. This is one of the reasons why a practical, fully automatic process planner is still not available. However, a trained human planner can do geometry analysis relatively easily. More details on geometric reasoning can be found in Chang (1990).

Stock Selection

Manufacturing is a shape-transformation process. Beginning with a stock material, a sequence of manufacturing processes is applied to transform the stock into the final shape. When the transfor- mation is done using machining, minimizing the volume of materials removed is desirable. Less material removal means less time spent in machining and less tool wear. The stock shape should be as close to the finished part geometry as possible. However, in addition to the minimum material removal rule, one also has to consider the difficulty of work holding. The stock material must be

clamped or chucked on the machine tool before cutting can be performed. This fixturing consideration has practical importance. However, raw materials can be purchased only in limited shapes and di- mensions. For example, steel is supplied in sheets of different thickness, length, and width, strips, wires, bars, and so on. The general shape has to be determined, and then stock preparation (cutting) has to be done to produce the stock. This is also true for the forming processes.

After the stock material is selected, one can compare the stock with the finished part and decide the volume of material to be removed for each process. In automated process planning, often the difference between the stock and the part, called the delta volume, is calculated. Geometric reasoning is performed on the delta volume. Without first selecting the stock, one cannot be certain exactly how to proceed with the manufacturing processes.

In some cases, especially for mass production, minimizing metal removal means preparing a casting as the stock material. Machining is used to improve critical or hard-to-cast surfaces. In this case, the casting has to be designed based on the part. Using casting as the stock minimizes the machining yet requires a high initial investment (casting design, mold making, etc.).

Gross Process Determination

Process planning can be separated into two stages: gross planning and detailed planning. Often only detailed planning is discussed in the process planning literature. Actually, the gross planning is even more critical than the detailed planning. Gross planning is used to determine the general approach to produce a part. For example, the same part geometry may be created through casting, machining, 3D fabrication, or welding of sheet metal. Only when a general approach is determined may one proceed with the detailed planning.

Casting, Machining, and Joining

One of the first decisions a process planner needs to make is whether to cast the part or machine it. Rough casting, such as sand casting, requires a good amount of machining. However, precision casting, such as die casting and investment casting, can produce almost net shape part (finished part). The decision is based on both economics and material properties; this issue will be addressed below.

In the initial evaluation, the technical feasibility and relative costs are taken into consideration. Process capabilities and relative cost for each casting and machining process are used in the evaluation. Based on the geometry to be created, material used, and production volume, one can propose several alter- native processes. The process capability includes shape capabilities and accuracy. Casting processes, such as die casting and investment casting, can create intricate geometry with relative high precision

(0.01 in.). However, sand casting is much worse. For tighter dimensional control, secondary machin- ing is a must.

Certain parts can be built using the welding process as well. This is especially true for structural parts. Large structural parts such as ship hulls are always built using welding. Medium-sized structural parts such as airplane structure parts may be joined together from machined parts. Sectional structures for jet fighters are always machined from a solid piece of metal in order to obtain the maximum strength. For small structures, all manufacturing methods are possible. When there are several alter- natives, an initial selection needs to be made.

Product Strength, Cost, etc.

Parts made using different stocks and processes exhibit different strength. As mentioned above, sectional structures for jet fighters are always machined from a solid piece of alloy steel. The raw material is homogeneous and rolled into the shape in the steel mill. It has high strength and consistent properties. When casting is used to form a part, a certain number of defects can be expected. The cooling and thus solidification of the material are not uniform. The surface area always solidifies first and at a much higher rate than the interior. A hard shell forms around the part. Depending on the complexity of the part, the type of mold used (sand, metal, etc.) and the mold design, voids, hot tear, cold shut, or other problems can happen. The part is not as strong as those produced using machining. In the case of welded parts, the welded join may not be as strong as the rest of the part.

As for the manufacturing cost, there is an even greater difference. For machining, the initial cost is low but the incremental cost is higher. The opposite is true for casting. Figure 6 shows the comparison. The initial cost for casting includes the cost of designing and building the mold and is relatively high. The slope of the incremental cost is the casting cost per piece. For machining, the initial cost is relatively much lower. On a manually controlled machine, only tools and fixtures need to be purchased. When a CNC machine is used, the programming cost has to be added to the fixed cost. However, the machining cost per piece will be lowered.

There are always alternative ways to make a part. Unless the way is specified by the designer, a good process planner always considers all possible alternatives and evaluate them. This evaluation need not always be carried out formally and precisely. Using past experience and with rough esti- mates, one can quickly eliminate most alternatives. The most promising alternatives have to be explored further before they are accepted or discarded.

Setup and Fixture Planning and Design

Let us assume that machining is the best alternative for the production of a part. Using the result of geometry analysis, we can group machining features based on their feasible approach directions. Most machining processes require the workpiece be positioned so that the tool orientation matches with the feature approach direction. For example, the part in Figure 7 consists of four holes. All holes have the same approach direction. Therefore, it is best to set up the workpiece with the holes aligned with the machine spindle. The position of the workpiece is called the setup position. When there are multiple approach directions, multiple setups may be needed to finish the machining op- erations.

A fixture is used to hold the workpiece at the desired setup position. For simple workpieces, a vise is sufficient to do the job. However, for more complex workpieces, locators and clamps are needed to position and clamp the workpiece on the machine table. After each setup the workpiece geometry is transformed (Figure 8). The finished part is obtained after the last setup is done. After a fixture is designed or configured (using modular figures), tool interference has to be checked. Obviously, fixture elements should not interfere with the tool motion. If the current fixture does interfere with the tool, either a new fixture needs to be designed or the machining feature that causes interference will not be cut during this setup. Again, this illustrates the iterative nature of process planning steps.

Process Selection

Process is defined as a specific type of manufacturing operation. Manufacturing processes can be classified as casting, forming, material-removal, and joining processes. Under casting are sand casting, investment casting, die casting, vacuum casting, centrifugal casting, inject molding, and so on. Form- ing includes rolling, forging, extrusion, drawing, powder metallurgy, thermoforming, spinning, and so on. Material removal includes milling, turning, drilling, broaching, sawing, filing, grinding, elec- trochemical machining (ECM), electrical-discharge machining (EDM), laser-beam machining, water- jet machining, ultrasonic machining, and so on. Joining processes include arc welding, electron-beam welding, ultrasonic welding, soldering, brazing, and so on. In this chapter only material-removal processes examples are used.

Each process has its own geometric and technological capabilities. Geometric capability means the shapes a process can create and the geometric constraints it might have. For example, the drilling process usually creates round holes and due to the drill nose cone the hole bottom has the cone shape. Technological capability includes tolerances (both dimensional and geometrical), surface finish, and surface integrity. Processes are selected based on the machining features in a setup. In the previous example (Figure 7), a setup consists of four holes. The hole geometry matches the drilling process capability. Therefore, drilling is selected for the holes. In this example, the drilling sequence has no effect on the final product or the manufacturing cost. However, in many other cases the process sequence does matter. For example, in Figure 9 there are four holes and a step. It makes more sense to mill the steps before drilling the holes than the other way around. When the two left- hand holes are drilled first, much of the drilling is done in vain. The milling process will remove the top half of the holes drilled. The process sequence is determined based on the relationship between features and the technological constraints of the processes. A good process planner takes all these into consideration when selecting the processes.

Process Detailing

Process detailing involves filling the details for the process selected. It includes determining the tool for the process, tool parameters (total length, diameter, cutting length, etc.), and process parameters (speed, feed, depth of cut, etc.).

Tool Selection

In order to carry out the process selected for a feature, a cutting tool is needed. Many different cutting tools can be used for the same process. In drilling, for example, there are different types of drill bites, such as twist drill, spade drill, and gun drill. Each drill also comes with different diameters, cutting length, total length, nose angle, and tool material. For drilling the holes in the example (Figure 9), two different drill lengths with the same diameter are needed. Of course, the longer drill can be used to drill shorter holes, too. However, if the diameters are slightly different, separate drills need to be specified.

Figure 10 shows different kinds of drills and turn tools. The selection of a tool depends on the feature geometry and geometric constraints. In considering milling, there are even more tool param- eters to consider. In addition to tool diameter, cutting depth, there are also such factors as number of cutting teeth, insert material, and rake angle. For end mills, there are also bottom-cutting and non- bottom-cutting types. Faced with this vast amount of choices, one must often rely on past experience and, for unfamiliar tools, handbooks.

Process Parameters Determination

Process parameters include speed, feed, and depth of cut. They are critical to the cutting process. Higher parameter values generate higher material-removal rate, but they also reduce the tool life. High feed also means rougher surface finish. In drilling and turning, feed is measured as how much the tool advances for each rotation of the tool. In milling, it is the individual tooth advancement for each tool rotation. In turning, for example, smaller feed means closely spaced tool paths on the part surface. The finish will be better in this case. Higher feed separates the tool paths and in the worst case creates uncut spacing between two passes. Types of process, the tool and workpiece materials, and hardness of the workpiece material affect process parameters. The parameter values are deter- mined through cutting experiments. They can be found in the tool vendor’s data sheets and in the Machining Data Handbook (Metcut 1980). These data are usually based on the constant tool life value, often 60 minutes of cutting time. When the required surface finish is high, one must modify the recommended parameters from the handbook. For new materials not included in any of the cutting parameter handbooks, one must conduct one’s own cutting experiments.

Process Optimization

It is always desirable to optimize the production. While global optimization is almost impossible to achieve, optimization on the process level is worth trying. The total production time is the sum of the cutting time, the material handling time, and the tool-change time. Shorter cutting time means faster speed and feed and thus shorter tool life. With a shorter tool life, the tool needs to be changed more frequently and the total time is thus increased. Because there is a trade-off between cutting time and tool life, one may find the optimal cutting parameters for minimum production time or cost. The techniques of process optimization are based on an objective function (time, cost, or another criterion) and a set of constraints (power, cutting force, surface finish, etc). The process-optimization models will be discussed later in the chapter. Finding optimal cutting parameters is fairly complex and requires precise data on both tool life model and machining model. Except in mass production, process optimization is generally not considered.

Plan Analysis and Evaluation

In on an old study conducted in industry, when several process planners were given the same part design, they all came up with quite different process plans. When adopted for the product, each

process plan resulted in different costs and part quality. If multiple process plans can be prepared for the same part, each plan must be analyzed and the best selected based on some preset criteria. If only one plan is prepared, it must be evaluated to ensure that the final result is acceptable. The final result here means the product quality.

Machining Time and Cost Estimation

Machining time can be calculated based on the cutting parameters and the feature geometry and dimension. It is used to estimate the production time and cost. It is also used in scheduling for determining the machine time. For turning and drilling, machining time can be calculated by the following formula:

Machining cost can be calculated by the machining time times a machine and operator overhead rate.

Estimated Product Quality

The commonly considered technological capabilities of a process include tolerances and surface finish. Surface finish is determined by the process and process parameters. It is affected not by the order in which processes are applied but only by the last process operated upon the feature. However, tolerances are results of a sequence of processes. Operation sequences will affect the final tolerance. Using a simple 2D part, Figure 10 shows the result of different process sequences. The arrow lines are dimension and tolerance lines. The drawing shows that the designer had specified dimensions and tolerances between features AB, BC, and CD. Notice that in this case features are vertical surfaces. If one uses feature A as the setup reference for machine B, B for C, and C for D, the produce part tolerance will be the same as the process tolerance. However, it would be tedious to do it this way. One may use A as the reference for cutting B, C, and D. In this case, tolerance on AB is the result of the process that cut B (from A to B). However, the tolerance on BC is the result of processes that cut feature B (from A to B) and feature C (from A to C). The finished tolerance on BC is twice the process tolerance and twice that of AB. The same can be said for CD. If we choose D as the reference, of course, the tolerance on CD is smaller than that for AB and BC. So we may conclude that process sequence does affect the quality of the part produced. The question is whether the current process sequence satisfies the designed tolerance requirements. Often this question is answered with the tolerance charting method. Tolerance charting will be introduced in the next section.