- Email: [email protected]
A hole-machining process planning system for marine engines
Journal of Manufacturing Systems 32 (2013) 114–123
Contents lists available at SciVerse ScienceDirect
Journal of Manufacturing Systems journal homepage: www.elsevier.com/locate/jmansys
Technical paper
A hole-machining process planning system for marine engines Cheol-Soo Lee a , Jae-Hyun Lee b , Dong-Soo Kim b , Eun-Young Heo c , Dong-Won Kim d,∗ a
Department of Mechanical Engineering, Sogang University, Seoul 121-742, South Korea Technology Research Center, CSCAM, 1235-10 Ok-dong, Gwangsan-gu, Gwangju, South Korea c Sogang Institute of Advanced Technology, Seoul 121-742, South Korea d Department of Industrial and Information Systems Engineering, Research Center for Industrial Technology, Chonbuk National University, Jeonju 561-756, South Korea b
a r t i c l e
i n f o
Article history: Received 22 May 2011 Received in revised form 20 October 2012 Accepted 29 October 2012 Available online 20 December 2012 Keywords: Computer-aided process planning Marine engine Hole-making Sequencing Process planning Multi-axis machining
a b s t r a c t This study proposes a hole-machining process planning system for marine engines, which converts industrial field requirements to the rules of the system. Unlike a fully automated system, the proposed system satisfies the requirements effectively by allowing the user to choose and to edit the rules. A computeraided process planning (CAPP) system is comprised of Hole Manager, Cutting Sequence Definition, and Operation Manager which are derived from the conventional knowledge based system. For the purpose of efficiently coordinating the system operations, a procedure is proposed as: (a) defining priorities for each operation, using properties for the nested cutting, the number of tool changes, the directions of the tool, the tool diameter, and the hole height, (b) calculating the score for each operation with the related priority level, and (c) sorting of operations by the score in an ascending order. This idea is quite simple but yields a significant efficiency along with a high flexibility. By changing the priority of elements, various operation sequences can be obtained. The proposed method also considers multi-axis machining and the use of special attachments. This paper describes the construction of a practical hole-making CAPP system that satisfies the specific requirements of marine engine machining. The applied examples are machined by using the proposed system, including an engine block and a cylinder header. © 2012 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
1. Introduction Computer-aided process planning (CAPP) is one of the most important steps in a manufacturing system. Various conditions and resources affect the total efficiency of the manufacturing system. The tool traveling and tool changing time comprises of a large portion among the total time in a manufacturing process while the optimization of these processes can contribute greatly to the total efficiency of a manufacturing system. Recently, many CAPP relevant studies have been carried out to improve productivity and the research scope has been expanded to the system level. For an illustration, the construction of a knowledge-based system and knowledge management in relationship with product lifecycle management (PLM) and enterprise resources planning (ERP) systems are embedded in CAPP. There has been significant increase in the number of researches and
Abbreviations: CAD, computer-aided design; CAM, computer-aided manufacturing; CAPP, computer-aided process planning; ERP, enterprise resources planning; MCS, machine coordination system; MEMS, microelectro-mechanical systems; PLM, product lifecycle management; PSO, particle swarm optimization; TSP, traveling salesman problem. ∗ Corresponding author. Tel.: +82 63 270 2328; fax: +82 63 270 2333. E-mail address: [email protected] (D.-W. Kim).
developments of the CAPP system being published recently. However, a viable off-the-shelf solution has not been commercialized due to the complexity of the CAPP system, and abundances of the interdependent requirements and conditions [1]. 1.1. Knowledge-based system overview Due to the needs of satisfying the requirements of industry, the knowledge-based system has received a great deal of attention from researchers. Park [2] proposed a knowledge base as a framework of process planning while customizable rules and a chosen methodology can be used for the control. The methodology has four knowledge elements: facts, constraints, modes of thinking, and rules, which are derived from a traditional three-phase modeling framework. Denkena et al. [1] suggested standard stages of a constructing process planner environment and a holistic PLM/CAPP solution construction methodology. In this approach, the ontology structure was used to make explicit knowledge contained within the system application. Once a knowledge base is constructed, the appropriate decisionmaking techniques are applied to explore the large solution space effectively under various constraints. These techniques adopted heuristic approaches, fuzzy logic, expert systems [3], reasoningbased approaches [4] and neural networks. Many studies have addressed the implementations of automated operation planning
0278-6125/$ – see front matter © 2012 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jmsy.2012.10.005
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
and tool selection algorithms. Generally, a Petri-net model has been regarded efficient in minimizing the number of tools and setup changes [5]. Branch and fathoming algorithms provided optimal and near optimal solutions for problems associated with the operation sequencing [6]. Simulated annealing algorithms optimized the selection and sequencing operations [7]. Relevant to the operation sequence and tool selection, heuristic combined knowledge bases were used [8–11]. Regarding the tool path planning, the heuristic approaches which are capable of providing near optimal solutions with relatively little computational effort have been widely proposed. These are Tabu search algorithms which minimize the total processing cost of tool traveling, tool changing, and the tool cutting time for hole-making operations [12]. A new approach based on particle swarm optimization (PSO) also has been introduced for drilling path optimizations [13]. An ant algorithm minimized the summation of the tool traveling and tool changing time [11]. On the basis of these decision techniques, there have been several attempts to construct practical CAPP systems. These include surface micro-machined MEMS devices [14], an activity model which defines sheet metal process planning [15], a feasible approach to the integration of CAD and CAPP [16], and an assembly/disassembly sequencing process [17,18]. Denkena et al. introduced a holistic process-planning model based on an integrated approach combining business and technological considerations [19]. Halevi and Wang introduced a road map method for flexibility and dynamics in the manufacturing process and thus simplified the decision-making process in production planning [20]. To meet the requirements of large variety of products in small batch sizes, Wang et al. proposed distributed process planning using function blocks [21,22]. In this research, function blocks can generate detailed and adaptive operation plans at runtime to best utilize the capability of the available machines, and the tasks of process planning are divided into two groups: shop-level supervisory planning and controller-level operation planning. Further, Wang et al. showed a hybrid approach using both knowledgebased and geometric reasoning rules in sequencing of interacting prismatic machining features [23]. Recently, he reviewed function block-based process planning and execution control systems [24]. The past and recent CAPP researches have been summarized in a number of categories by Xu et al. [25], i.e. feature-based technologies, knowledge-based systems, artificial neural networks, genetic algorithms, fuzzy set theory and fuzzy logic, Petri nets, agent-based technology, Internet-based technology, STEP-compliant CAPP and other emerging technologies.
115
Fig. 1. The manufacturing process of a marine engine.
The ultimate objective of a CAPP system is to reduce the degree of human judgment in the process planning. The CAPP system of the above techniques, however, has a limitation in a direct application to the industrial field. Furthermore, the construction of a new CAPP system from the ground up requires a considerable amount of time. In this paper, thus, a hole-making system which has been developed with the knowledge from a domain expert is embedded into a commercial CAD system. 1.2. Proposed system Fig. 1 shows the manufacturing process of a marine engine where the final product consists of several parts. The assembled faces of the parts are machined by face-milling and hole-making prior to the assembly. The holes serve to hold the parts together and they are designed to endure high temperatures and high pressures.
Fig. 2. Some examples of attachments to machine holes.
116
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
To satisfy this machining environment, this paper presents a novel methodology with four steps which are outlined as follows: (1) Definition of the requirements: the requirements in the industrial field are complex and interdependent. Therefore, definition of the requirements holds a high importance in the construction of a system. (2) Design of a system framework: the system process planner designs the structure of the main system with a reflection of the domain knowledge from the expert. Fig. 3 shows the proposed system framework. (3) Definition of a rule: to satisfy the requirements, rules are defined and stored in the modules, or parts of the system. A rule is a logical procedure to search for an optimum solution in each system module. A rule can be related to several requirements, as requirements can be interrelated. (4) Selection of preference rules: for the global efficiency of machining, the system user can select the preferred rules or edit the interim findings to fit his intentions. 2. Hole-making system of a marine engine
Fig. 3. The proposed system framework.
2.1. System overview They are often located in deep and narrow area of an assembly part. To machine these holes, attachments are used as shown in Fig. 2. The parts are large and expensive in general and there are special requirements related to the sequence planning of the operation such as tool path planning and handling of machining errors. These factors should be reflected in CAPP system for machining marine engine block. In case of the present manual system, each milling path and hole making condition is defined by concrete user operations. The recent fully automated commercial CAD/CAM systems do not allow users to apply field rules. From the efficiency perspective, the need for developing a system which is capable of satisfying the manufacturing conditions on the base of commercial system still persists.
The system framework consists of sub modules with rules reflecting requirements. Unlike a fully automated system, the proposed system is capable of reflecting a specific requirement by allowing the user to choose and edit the rules. As shown in Fig. 4, the system consists of Hole Manager, Cutting Sequence Definition and Operation Manager. An efficient sequence of operations can be obtained for a three-axis machine, as well as a multi-axis machine. 2.2. CAD model and feature extraction A part can be designed according to its features using a solid modeler. Thus, a hole feature can automatically extracted and the
Fig. 4. Hole-making system of a marine engine.
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
117
Fig. 5. Tool parameter definition.
tools for cutting a hole can be optimally sequenced via process planning. Even though many studies have been carried out for feature extraction techniques [26,27], the modeling of all the complex hole features can be a time-consuming work to the CAM programmer if a CAD/CAM division is not well organized. Furthermore, the feature extraction is not straightforward. In case of a complex hole, a user creates a simple drill hole rather than a fully featured hole in the previously proposed systems. The hole shape is defined by setting the tool parameters at the tool definition stage of Hole Manager. As for the revision management, a pre-worked part should be reusable for efficiency because it is likely that there are many similar parts existing in a marine engine. However, a commercial solid
modeling system does not support an individually maintained hole identifier (hole ID) when a part is reloaded. By saving the hole ID at the attribute field of the hole feature, the hole features are managed individually. The revision can be performed with less effort. 2.3. Hole Manager 2.3.1. Definition of tool In this module, cutting tools are selected from the tool DB and the machining parameters of each tool are determined while the selected tools are sequenced to satisfy the precedence constraints
Fig. 6. Sample engine block.
118
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
Table 1 Tool definition and hole group. Hole
1 2 3 4 Hole group
Hole A
Table 2 Tool priority. Hole B
Hole C
Tool
Dia.
Tool
Dia.
Tool
Dia.
C-drill Drill Drill
0 18 30
C-drill Drill Bore Tap
0 18 30 20
C-drill Drill Chamfer Tap
0 18 40 20
Hole A
Hole B
Hole C
HoleA G1 HoleA G2 HoleA G3
HoleB G1
HoleC G1
[1–3]. A feasible plan for the hole-making [16], for illustration, can be described as: (a) (b) (c) (d)
hole starting (center drilling); core making (twist/spade/gun drilling)’ hole improving (reaming/boring/precision boring); and hole finishing (grinding/honing).
The selection of the tools and the machining parameters are interactively determined by an experienced worker. For a complex hole, defining machining parameters such as the cutting depth, the length of engagement and retraction precisely, which are important to define the operation can be a challenging task. To assist with this process, when the user defines the depth value of each tool, the machined shape of a hole and grid lines permitted for the selected depth are displayed interactively as shown in Fig. 5. In this manner, a user can visually validate whether the stock is adequately machined. 2.3.2. Grouping of hole After the tool definition, the holes are selected and categorized into groups (hole group) according to the hole direction, the hole height, and the distance from the specific position. The hole groups are added to the data storage of Hole Manager. 2.4. Cutting Sequence Definition In this stage, the operations are sequenced by five elements (nested cutting, tool number, hole group, tool diameter, and hole height). The tool path is then optimized. Finally, to support the practical requirements of a marine engine, additional processes are considered. 2.4.1. Operation sequence planning To explain the sequencing of an operation, a marine engine block is introduced. Fig. 6 shows the engine block and Table 1 shows the initial conditions of the part. There are three types of hole feature: simple drill (Hole A), counter bore (Hole B) and counter sink (Hole C). Hole A is classified into three groups according to the hole direction and hole height; Hole A G1, Hole A G2 and Hole A G3. Hole B and Hole C consist of one hole group. Among the hole groups in the +Z hole direction, the height of Hole B is higher than that of the other two hole groups of Hole A G1 and Hole C. There are 17 operations in total: 4 operations for Hole B, 4 operations for Hole C, and 9 operations for the operations of Hole A (three hole groups multiplied by three tools). The operation sequences are planned by following these three steps: (a) definition of the priorities for the sequence element (b) calculation of the score of each operation using the priorities, and
Tool number
Tool
Tool priority
#1 #2 #3 #4 #5 #6
C-drill Drill 18 Drill 50 Chamfer 40 Bore 30 Tap M20 × 2.5
1 2 3 5 4 6
Table 3 Hole group priority. Index
Hole group
Tool direction
Group priority
1 2 3 4 5
HoleA G1 HoleB G1 HoleC G1 HoleA G2 HoleA G3
(0, 0, 1) (0, 0, 1) (0, 0, 1) (0,-1, 0) (1, 0, 0)
1 1 1 2 3
(c) sorting of the operations by score in an increasing order. Using these steps, a feasible solution of the operation sequence is calculated. The details for each step are as follows: (1) Priorities for the sequencing The types of priority for the sequencing are listed below: • Nested cutting: If the hole-making process has completed from the first tool to the final tool, it is categorized to nested cutting and placed in the preceding position. • Tool number: In Table 2, a sample priority for tool number is defined. This element categorizes the operations according to the number of tools while operations need to minimize the number of tool changes. • Hole group (tool direction): The priority of hole group is illustrated in Table 3. The holes with a same direction tend to have an identical value and the planner selects the order of the machining directions. • Tool diameter: Tool priorities are determined in the ascending order of tool size as shown in Table 2. • Hole height: As shown in Fig. 6, holes of the Hole B G1 group are deeper than those of the Hole A G1 and HoleC G1 groups. Therefore, holes of the Hole B G1 group are set to the reference height. The height of these holes is set to zero point and the remaining groups are ordered by their heights. (2) Calculation of the score and determination of the sequence The operation sequence is influenced by the priority of the sequence element. If the priority of the sequence element is well defined, operations can be arranged to satisfy the requirements of the user. The score of individual operation is calculated as a function of the priority of the sequence element. The score of each operation is calculated as; Score = −En.c. × N 4 + Et.p. × N 3 + Eh.g. × N 2 + Et.d. × N 1 + Eh.h. × N 0
(1)
where En.c., the element of the nested cutting; Et.p., the element of the tool priority; Eh.g., the element of the hole group priority; Et.d., the element of the hole diameter; Eh.h., the element of the hole height; Mn.c., the max value of the nested cutting property; Mt.p. = the max value of the tool priority; Mh.g., the max value of the hole group priority; Mt.d., the max value of the hole diameter; Mh.h., the max value of the hole height; and N, max (Mn.c., Mt.p., Mh.g., Mt.d., Mh.h.).
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
119
Table 4 Examples of the operation sequence. ID
Hole group
Tool
Priority rule
Tool number
Hole group
Tool diameter
Hole height
3
2
1
0
(a) Case 1 1 10 14 4 7 2 11 15 5 8 3 6 9 12 13 16 17
HoleA G1 HoleB G1 HoleC G1 HoleA G2 HoleA G3 HoleA G1 HoleB G1 HoleC G1 HoleA G2 HoleA G3 HoleA G1 HoleA G2 HoleA G3 HoleB G1 HoleB G1 HoleC G1 HoleC G1
C-drill C-drill C-drill C-drill C-drill Drill 18 Drill 18 Drill 18 Drill 18 Drill 18 Drill 30 Drill 30 Drill 30 Bore 30 Tap 20 Chamf. 40 Tap 20
1 1 1 1 1 2 2 2 2 2 3 3 3 4 5 5 6
1 1 1 2 3 1 1 1 2 3 1 2 3 1 1 1 1
0 0 0 0 0 18 18 18 18 18 30 30 30 30 20 40 20
0 0 30 0 30 0 0 30 0 30 0 0 30 0 0 30 30
(b) Case 2 1 10 14 2 11 16 3 12 13 16 17 4 5 6 7 8 9
HoleA G1 HoleB G1 HoleC G1 HoleA G1 HoleB G1 HoleC G1 HoleA G1 HoleB G1 HoleB G1 HoleC G1 HoleC G1 HoleA G2 HoleA G2 HoleA G2 HoleA G3 HoleA G3 HoleA G3
C-drill C-drill C-drill Drill 18 Drill 18 Drill 18 Drill 18 Bore 30 Tap 20 Chamf. 40 Tap 20 C-drill Drill 18 Drill 30 C-drill Drill 18 Drill 30
1 1 1 1 1 1 1 1 1 1 1 2 2 2 3 3 3
1 1 1 2 2 2 3 4 5 5 6 1 2 3 1 2 3
0 0 0 18 18 18 30 30 20 40 20 0 18 30 0 18 30
0 0 30 0 0 30 0 0 0 30 30 0 0 0 30 30 30
(c) Case 3 1 10 2 11 3 12 13 14 15 16 17 4 5 6 7 8 9
HoleA G1 HoleB G1 HoleA G1 HoleB G1 HoleA G1 HoleB G1 HoleB G1 HoleC G1 HoleC G1 HoleC G1 HoleC G1 HoleA G2 HoleA G2 HoleA G2 HoleA G3 HoleA G3 HoleA G3
C-drill C-drill Drill 18 Drill 18 Drill 30 Bore 30 Tap 20 C-drill Drill 18 Chamf. 40 Tap 20 C-drill Drill 18 Drill 30 C-drill Drill 18 Drill 30
1 1 1 1 1 1 1 1 1 1 1 2 2 2 3 3 3
0 0 0 0 0 0 0 30 30 30 30 0 0 0 30 30 30
1 1 2 2 3 4 5 1 2 5 6 1 2 3 1 2 3
0 0 18 18 30 30 20 0 18 40 20 0 18 30 0 18 30
N is the weight value to avoid interferences and is the maximum among sequence elements. Generally, when N is 1000, there are no interferences. The feasible cases of the elements priority are: Case 1: classification by the number of tool changes, Case 2: classification by the tool direction, and Case 3: classification by the hole height of the same tool direction.
priority for tool number is higher than the priority for hole group. The score of ID = 1 (HoleA G1 Center Drill) for Case 1 in Table 4 is calculated as: 1, 001, 000, 000 = −0 × 10004 + 1 × 10003 + 1 × 10002 + 0 × 10001
(2)
The discussion for each case is as follows: Case 1. Classification by the number of tool changes The objective of Case 1 is to minimize the number of tool changes. The priorities (nested cutting, tool number, hole group, tool diameter, and hole height) are given as 4, 3, 2, 1, and 0. The
Table 4(a) shows the classified operations. Tool Number has the first order and Hole Group has the second order. Case 2: classification by the tool direction, and This case classifies the operations with the hole direction. The priorities are given as 4, 2, 3, 1 and 0. Hole group has a higher
120
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
Fig. 7. Multi-axis machine: (a) frame box of an engine block and (b) cylinder header.
order than tool number. This case can be useful when the cutting direction does not change. Case 3: classification by the hole height of the same tool direction. In this case, the priorities are given as 4, 1, 3, 0 and 2. To classify the holes with same heights and tool diameters, the priority elements are set as: hole group = 3 and hole heigh t = 2. The operations with zero height precede operations of the height of 30 as shown in Table 4(c). Nested cutting element has the highest priority. In the score function, the value of nested cutting element is negative. Accordingly, the operation checks whether nested cutting element is located at the first position of the operation list. The other values are all positive. If nested cutting elements for HoleA G1 operations have a value of 1 and if those for HoleA G3 have a value of 2, the operations of HoleA G3 are executed first. After that, the operations of HoleA G1 are executed as shown in Table 4(a). The proposed system allows a user to edit the priorities under the consideration of the tool sequence constraints defined in tool definition. In Case 1 of Table 4, the 6th operation (ID = 2, Drill 18 of Hole A-G1) is interchangeable with operations ranging from the 1st operation (ID = 1, C-drill of Hole A-G1) to the 11th operation (ID = 2, Drill 30 of Hole A-G1).
(1) Prismatic part A prismatic part of an engine block is huge in its size while the height can be as much as 4–5 m and length may be more than 10 m; moreover, holes may exist in deep and narrow places. An attachment enables access to these holes and to machine the holes in all directions except for the bottom direction. The hole machining method is similar to the conventional 2D drilling except for the tool axis changing. To support the tool and attachment changing, an additional tool path trajectory and attachment change codes are added. The additional tool path trajectory should be calculated for avoiding collisions between the moving objects (the attachment and tool) and the work piece. The attachment information of each machine is stored in the system database in advance. Thus, the number of the suitable attachments and the related parameters can be easily retrieved and inserted as shown in Fig. 7(a). (2) Cylindrical part
2.4.2. Tool path planning The tool path can be optimized by the traveling salesman problem (TSP) approach [5–8]. In other cases, holes should be categorized by the specific direction or the hole height. As shown in Table 4(a), the 1st and 2nd operations share the same tool number and height therefore these two operations are combined. The y-axis element of the distance vector of two points is higher weighted than the x-axis element of that in the TSP. In the case of many holes, the holes are pre-grouped by direction. Each group is considered as one hole in the TSP. In this way, the tool path can be planned in the x direction preferred rule. 2.4.3. Multi-axis machining To machine a part of a marine engine, multi-axis machining is essential. The shape of the part, as shown in Fig. 7, can be prismatic or cylindrical, and are described below.
Fig. 8. Interference checking.
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
121
Fig. 9. Main system flow for an engine block.
A cylindrical part is machined by a four-axis or five-axis machine with a turning table. The position and the orientation vector of a hole are calculated on the basis of a machine coordination system (MCS). If the holes resemble the spokes of a wheel, the MCS should be defined at the center of the circularpatterned holes as shown in Fig. 7(b) before generating NC data. Hence, the local MCS is defined at the center of the circularpatterned holes. The system generates an NC code based on the local MCS. 2.4.4. Check of error and interference A marine engine is heavy and expensive in nature therefore the number of errors must be minimized during the machining operations. When the tool direction and the tool number change, the
tool path and the cutting operations need to be validated. One of the challenges in tool path generation is to generate interferencefree tool paths. A cutter and stock interference takes place at the point-to-point and operation-to-operation links. As to the pointto-point linking, if there is an interference found between the stock and the cutter, the parameters of the engagement and the retraction should be recalculated by the following steps. Firstly, an interference-free height between the stock and the tool path is calculated, and then the offset height considering cutter and attachment movements is added to that height as shown in Fig. 8(a). As to the operation-to-operation linking, an additional path to escape collisions is calculated. Generally, passage point of path is calculated by the extreme value of the stock boundary box. Alternatively, it can be edited by user-defined position as shown in Fig. 8(b).
122
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
Fig. 10. (a) Engine block and (b) operation list.
Fig. 11. (a) Cylinder header and (b) MCS and job name.
2.5. Operation Manager Operation Manager handles all of the different types of operations, including milling, turning, hole-making as well as the post-process information. This enables user to simulate, verify and check using various criteria. Operation Manager also allows operation editing. 3. Implementation result With the proposed concept, a hole-making system has been developed and Fig. 9 shows the main system flow with an engine block sample. The sample is selected for a case study, but the more complicated shapes of engine parts are commonly found in the industrial field. Fig. 9(b) is the main dialog, which can select and manage hole-types registered for machining beforehand. At each hole-type, the machining methods and rules are defined and hole-type instances are selected, as shown in Fig. 9(c). Then, the proper machining sequences are calculated (Fig. 9(d)) according to the cutting priority rules, and the results can be edited with easy, if additional conditions are changed. The final results are registered to the operation manager in Fig. 9(e). In the previous system, the machining sequences are defined manually in the operation manger step, thus it requires many hours to create and edit the machining data. For efficiency, however, the proposed system supports additional modules to handle previous machining data, as well as to secure the flexibility in machining sequence determination. Compared to the previous commercial system, the increased efficiency of the proposed system has been reported more than 60% in the actual industrial field. 3.1. Engine block Fig. 10(a) shows a machined engine block. The machined area was cut by a face-milling operation and all of the holes were machined using the proposed method. The operation sequence is categorized by the tool number, as shown in Fig. 10(b). 3.2. Cylinder header The cylinder header was machined by the four-axis and five-axis machines, as shown in Fig. 11(a). A cylindrical coordinate system
was applied to the circular-patterned holes and then the nested cutting was applied to the complex holes. In the operation list as shown in Fig. 11(b), MCS and its NC code (G54, G55, etc.) were added to each operation of the post-processing step. 4. Conclusions In this paper, a hole-making CAPP system is proposed to machine the parts of a marine engine. It consists of Hole Manager, Cutting Sequence Definition and Operation Manager. These components are derived from a conventional system. To satisfy the requirements, rules are defined for each module. For the operation sequence planning, three steps are proposed: (a) definition of sequence elements such as nested cutting, the tool number, the tool direction, the tool diameter, and the hole height, (b) score calculations for each operation, and (c) sorting of the operations by scores in an ascending order. By changing the priority of elements, the operation sequence can be adjusted. This idea is relatively simple but it is capable of yielding a significant efficiency along with a high flexibility. The proposed method was implemented in the commercial CAD/CAM system and utilized in industrial applications. Conflict of interest The authors have no conflict of interest. Acknowledgement This research was partly supported Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2011-0022044). References [1] Denkena B, Shpitalni M, Kowalski P, Molcho G, Zipori Y. Knowledge management in process planning. Annals of the CIRP 2007;56(1):175–80. [2] Park SC. Knowledge capturing methodology in process planning. ComputerAided Design 2003;35:1109–17. [3] Wong TN, Chan LCF, Lau HCW. Machining process sequencing with fuzzy expert system and genetic algorithms. Engineering with Computers 2003;19:191–202.
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123 [4] Yongtao H, Jingying M. A knowledge-based auto-reasoning methodology in hole-machining process planning. Computers in Industry 2006;57(4):297–304. [5] Kiritsis D, Neuendorf KP, Xirouchakis P. Petri-net techniques for process planning cost estimation. Annals of the CIRP 1998;47(1):427–30. [6] Lee DH, Kiritsis D, Xirouchakis P. Branch and fathoming algorithms for operation sequencing in process planning. Internal Journal of Production Research 2001;39(1):1649–69. [7] Lee DH, Kiritsis P, Xirouchakis P. Simulated annealing algorithms for operation sequencing in nonlinear process planning. Journal of the Korean Institute of Industrial Engineers 2001;27(3):315–27. [8] Lee CS, Kim SS, Choi JS. Operation sequence and tool selection in flexible manufacturing systems under dynamic tool allocation. Computers and Industrial Engineering 2003;45:61–73. [9] Etienne A, Dantan JY, Siadat A, Martin P. An improved approach for automatic process plan generation of complex borings. Computers in Industry 2006;57(7):663–75. [10] Kalahan F, Liang M. Hybrid genetic algorithm and simulated annealing approach for the optimization of process plans for prismatic parts. The International Journal of Product and Research 2002;40(8):899–922. [11] Ghaiebi H, Solimanpur M. An ant algorithm for optimization of hole-making operations. Computers and Industrial Engineering 2007;52(2):308–19. [12] Kalahan F, Liang M. A tabu search approach to optimization of drilling operation. Computers and Industrial Engineering 1996;31(1/2):371–4. [13] Onwubolu GC, Clerk M. Optimal path for automated drilling operations by a new heuristic approach using particle swarm optimization. International Journal of Production Research 2004;42(3):473–91. [14] Li J, Gao S, Liu Y. Solid-based CAPP for surface micromachined MEMS devices. Computer-Aided Design 2007;39(3):190–201. [15] Smith S, Zamudia C. Activity model and computer aided system for defining sheet metal process planning. Journal of Materials Processing Technology 2006;173:213–22.
123
[16] Zhou X, Qiu Y, Hua G, Wang H, Ruan X. A feasible approach to the integration of CAD and CAPP. Computer-Aided Design 2007;39(4):324–38. [17] Park HS. A knowledge-based system for assembly sequence planning. International Journal of Precision Engineering and Manufacturing 2005;1(2): 35–42. [18] Park HS, Choi HW. Disassembly process planning of end-of-life car. International Journal of Precision Engineering and Manufacturing 2005;6(1): 42–50. [19] Denkena B, et al. Knowledge management in process planning. CIRP Annals – Manufacturing Technology 2007;56(1):175–80. [20] Halevi G, Wang K. Knowledge based manufacturing system (KBMS). Journal of Intelligent Manufacturing 2007;18(4):467–74. [21] Wang L, Feng HY, Cai N. Architecture design for distributed process planning. Journal of Manufacturing Systems 2003;22(2):99–115. [22] Liu Z, Wang L. Sequencing of interacting prismatic machining features for process planning. Computers in Industry 2007;58(4):295–303. [23] Wang L, Cai N, Feng HY, Liu Z. Enriched machining feature based reasoning for generic machining process sequencing. International Journal of Production Research 2006;44(8):1479–501. [24] Wang L, Adamson G, Holm M, Moore P. A review of function blocks for process planning and control of manufacturing equipment. Journal of Manufacturing Systems 2012;31(3):269–79. [25] Xu X, Wang L, Newman ST. Computer-aided process planning – a critical review of recent developments and future trends. International Journal of Computer Integrated Manufacturing 2011;24(1):1–31. [26] Han J, Pratt M, Regli WC. Manufacturing feature recognition from solid models: a status report. IEEE Transactions on Robotics and Automation 2000;19(6):782–96. [27] Joshi N, Dutta D. Feature simplification techniques for freeform surface models. Journal of Computing and Information Science in Engineering 2003;3(9):177–86.
Contents lists available at SciVerse ScienceDirect
Journal of Manufacturing Systems journal homepage: www.elsevier.com/locate/jmansys
Technical paper
A hole-machining process planning system for marine engines Cheol-Soo Lee a , Jae-Hyun Lee b , Dong-Soo Kim b , Eun-Young Heo c , Dong-Won Kim d,∗ a
Department of Mechanical Engineering, Sogang University, Seoul 121-742, South Korea Technology Research Center, CSCAM, 1235-10 Ok-dong, Gwangsan-gu, Gwangju, South Korea c Sogang Institute of Advanced Technology, Seoul 121-742, South Korea d Department of Industrial and Information Systems Engineering, Research Center for Industrial Technology, Chonbuk National University, Jeonju 561-756, South Korea b
a r t i c l e
i n f o
Article history: Received 22 May 2011 Received in revised form 20 October 2012 Accepted 29 October 2012 Available online 20 December 2012 Keywords: Computer-aided process planning Marine engine Hole-making Sequencing Process planning Multi-axis machining
a b s t r a c t This study proposes a hole-machining process planning system for marine engines, which converts industrial field requirements to the rules of the system. Unlike a fully automated system, the proposed system satisfies the requirements effectively by allowing the user to choose and to edit the rules. A computeraided process planning (CAPP) system is comprised of Hole Manager, Cutting Sequence Definition, and Operation Manager which are derived from the conventional knowledge based system. For the purpose of efficiently coordinating the system operations, a procedure is proposed as: (a) defining priorities for each operation, using properties for the nested cutting, the number of tool changes, the directions of the tool, the tool diameter, and the hole height, (b) calculating the score for each operation with the related priority level, and (c) sorting of operations by the score in an ascending order. This idea is quite simple but yields a significant efficiency along with a high flexibility. By changing the priority of elements, various operation sequences can be obtained. The proposed method also considers multi-axis machining and the use of special attachments. This paper describes the construction of a practical hole-making CAPP system that satisfies the specific requirements of marine engine machining. The applied examples are machined by using the proposed system, including an engine block and a cylinder header. © 2012 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
1. Introduction Computer-aided process planning (CAPP) is one of the most important steps in a manufacturing system. Various conditions and resources affect the total efficiency of the manufacturing system. The tool traveling and tool changing time comprises of a large portion among the total time in a manufacturing process while the optimization of these processes can contribute greatly to the total efficiency of a manufacturing system. Recently, many CAPP relevant studies have been carried out to improve productivity and the research scope has been expanded to the system level. For an illustration, the construction of a knowledge-based system and knowledge management in relationship with product lifecycle management (PLM) and enterprise resources planning (ERP) systems are embedded in CAPP. There has been significant increase in the number of researches and
Abbreviations: CAD, computer-aided design; CAM, computer-aided manufacturing; CAPP, computer-aided process planning; ERP, enterprise resources planning; MCS, machine coordination system; MEMS, microelectro-mechanical systems; PLM, product lifecycle management; PSO, particle swarm optimization; TSP, traveling salesman problem. ∗ Corresponding author. Tel.: +82 63 270 2328; fax: +82 63 270 2333. E-mail address: [email protected] (D.-W. Kim).
developments of the CAPP system being published recently. However, a viable off-the-shelf solution has not been commercialized due to the complexity of the CAPP system, and abundances of the interdependent requirements and conditions [1]. 1.1. Knowledge-based system overview Due to the needs of satisfying the requirements of industry, the knowledge-based system has received a great deal of attention from researchers. Park [2] proposed a knowledge base as a framework of process planning while customizable rules and a chosen methodology can be used for the control. The methodology has four knowledge elements: facts, constraints, modes of thinking, and rules, which are derived from a traditional three-phase modeling framework. Denkena et al. [1] suggested standard stages of a constructing process planner environment and a holistic PLM/CAPP solution construction methodology. In this approach, the ontology structure was used to make explicit knowledge contained within the system application. Once a knowledge base is constructed, the appropriate decisionmaking techniques are applied to explore the large solution space effectively under various constraints. These techniques adopted heuristic approaches, fuzzy logic, expert systems [3], reasoningbased approaches [4] and neural networks. Many studies have addressed the implementations of automated operation planning
0278-6125/$ – see front matter © 2012 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jmsy.2012.10.005
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
and tool selection algorithms. Generally, a Petri-net model has been regarded efficient in minimizing the number of tools and setup changes [5]. Branch and fathoming algorithms provided optimal and near optimal solutions for problems associated with the operation sequencing [6]. Simulated annealing algorithms optimized the selection and sequencing operations [7]. Relevant to the operation sequence and tool selection, heuristic combined knowledge bases were used [8–11]. Regarding the tool path planning, the heuristic approaches which are capable of providing near optimal solutions with relatively little computational effort have been widely proposed. These are Tabu search algorithms which minimize the total processing cost of tool traveling, tool changing, and the tool cutting time for hole-making operations [12]. A new approach based on particle swarm optimization (PSO) also has been introduced for drilling path optimizations [13]. An ant algorithm minimized the summation of the tool traveling and tool changing time [11]. On the basis of these decision techniques, there have been several attempts to construct practical CAPP systems. These include surface micro-machined MEMS devices [14], an activity model which defines sheet metal process planning [15], a feasible approach to the integration of CAD and CAPP [16], and an assembly/disassembly sequencing process [17,18]. Denkena et al. introduced a holistic process-planning model based on an integrated approach combining business and technological considerations [19]. Halevi and Wang introduced a road map method for flexibility and dynamics in the manufacturing process and thus simplified the decision-making process in production planning [20]. To meet the requirements of large variety of products in small batch sizes, Wang et al. proposed distributed process planning using function blocks [21,22]. In this research, function blocks can generate detailed and adaptive operation plans at runtime to best utilize the capability of the available machines, and the tasks of process planning are divided into two groups: shop-level supervisory planning and controller-level operation planning. Further, Wang et al. showed a hybrid approach using both knowledgebased and geometric reasoning rules in sequencing of interacting prismatic machining features [23]. Recently, he reviewed function block-based process planning and execution control systems [24]. The past and recent CAPP researches have been summarized in a number of categories by Xu et al. [25], i.e. feature-based technologies, knowledge-based systems, artificial neural networks, genetic algorithms, fuzzy set theory and fuzzy logic, Petri nets, agent-based technology, Internet-based technology, STEP-compliant CAPP and other emerging technologies.
115
Fig. 1. The manufacturing process of a marine engine.
The ultimate objective of a CAPP system is to reduce the degree of human judgment in the process planning. The CAPP system of the above techniques, however, has a limitation in a direct application to the industrial field. Furthermore, the construction of a new CAPP system from the ground up requires a considerable amount of time. In this paper, thus, a hole-making system which has been developed with the knowledge from a domain expert is embedded into a commercial CAD system. 1.2. Proposed system Fig. 1 shows the manufacturing process of a marine engine where the final product consists of several parts. The assembled faces of the parts are machined by face-milling and hole-making prior to the assembly. The holes serve to hold the parts together and they are designed to endure high temperatures and high pressures.
Fig. 2. Some examples of attachments to machine holes.
116
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
To satisfy this machining environment, this paper presents a novel methodology with four steps which are outlined as follows: (1) Definition of the requirements: the requirements in the industrial field are complex and interdependent. Therefore, definition of the requirements holds a high importance in the construction of a system. (2) Design of a system framework: the system process planner designs the structure of the main system with a reflection of the domain knowledge from the expert. Fig. 3 shows the proposed system framework. (3) Definition of a rule: to satisfy the requirements, rules are defined and stored in the modules, or parts of the system. A rule is a logical procedure to search for an optimum solution in each system module. A rule can be related to several requirements, as requirements can be interrelated. (4) Selection of preference rules: for the global efficiency of machining, the system user can select the preferred rules or edit the interim findings to fit his intentions. 2. Hole-making system of a marine engine
Fig. 3. The proposed system framework.
2.1. System overview They are often located in deep and narrow area of an assembly part. To machine these holes, attachments are used as shown in Fig. 2. The parts are large and expensive in general and there are special requirements related to the sequence planning of the operation such as tool path planning and handling of machining errors. These factors should be reflected in CAPP system for machining marine engine block. In case of the present manual system, each milling path and hole making condition is defined by concrete user operations. The recent fully automated commercial CAD/CAM systems do not allow users to apply field rules. From the efficiency perspective, the need for developing a system which is capable of satisfying the manufacturing conditions on the base of commercial system still persists.
The system framework consists of sub modules with rules reflecting requirements. Unlike a fully automated system, the proposed system is capable of reflecting a specific requirement by allowing the user to choose and edit the rules. As shown in Fig. 4, the system consists of Hole Manager, Cutting Sequence Definition and Operation Manager. An efficient sequence of operations can be obtained for a three-axis machine, as well as a multi-axis machine. 2.2. CAD model and feature extraction A part can be designed according to its features using a solid modeler. Thus, a hole feature can automatically extracted and the
Fig. 4. Hole-making system of a marine engine.
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
117
Fig. 5. Tool parameter definition.
tools for cutting a hole can be optimally sequenced via process planning. Even though many studies have been carried out for feature extraction techniques [26,27], the modeling of all the complex hole features can be a time-consuming work to the CAM programmer if a CAD/CAM division is not well organized. Furthermore, the feature extraction is not straightforward. In case of a complex hole, a user creates a simple drill hole rather than a fully featured hole in the previously proposed systems. The hole shape is defined by setting the tool parameters at the tool definition stage of Hole Manager. As for the revision management, a pre-worked part should be reusable for efficiency because it is likely that there are many similar parts existing in a marine engine. However, a commercial solid
modeling system does not support an individually maintained hole identifier (hole ID) when a part is reloaded. By saving the hole ID at the attribute field of the hole feature, the hole features are managed individually. The revision can be performed with less effort. 2.3. Hole Manager 2.3.1. Definition of tool In this module, cutting tools are selected from the tool DB and the machining parameters of each tool are determined while the selected tools are sequenced to satisfy the precedence constraints
Fig. 6. Sample engine block.
118
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
Table 1 Tool definition and hole group. Hole
1 2 3 4 Hole group
Hole A
Table 2 Tool priority. Hole B
Hole C
Tool
Dia.
Tool
Dia.
Tool
Dia.
C-drill Drill Drill
0 18 30
C-drill Drill Bore Tap
0 18 30 20
C-drill Drill Chamfer Tap
0 18 40 20
Hole A
Hole B
Hole C
HoleA G1 HoleA G2 HoleA G3
HoleB G1
HoleC G1
[1–3]. A feasible plan for the hole-making [16], for illustration, can be described as: (a) (b) (c) (d)
hole starting (center drilling); core making (twist/spade/gun drilling)’ hole improving (reaming/boring/precision boring); and hole finishing (grinding/honing).
The selection of the tools and the machining parameters are interactively determined by an experienced worker. For a complex hole, defining machining parameters such as the cutting depth, the length of engagement and retraction precisely, which are important to define the operation can be a challenging task. To assist with this process, when the user defines the depth value of each tool, the machined shape of a hole and grid lines permitted for the selected depth are displayed interactively as shown in Fig. 5. In this manner, a user can visually validate whether the stock is adequately machined. 2.3.2. Grouping of hole After the tool definition, the holes are selected and categorized into groups (hole group) according to the hole direction, the hole height, and the distance from the specific position. The hole groups are added to the data storage of Hole Manager. 2.4. Cutting Sequence Definition In this stage, the operations are sequenced by five elements (nested cutting, tool number, hole group, tool diameter, and hole height). The tool path is then optimized. Finally, to support the practical requirements of a marine engine, additional processes are considered. 2.4.1. Operation sequence planning To explain the sequencing of an operation, a marine engine block is introduced. Fig. 6 shows the engine block and Table 1 shows the initial conditions of the part. There are three types of hole feature: simple drill (Hole A), counter bore (Hole B) and counter sink (Hole C). Hole A is classified into three groups according to the hole direction and hole height; Hole A G1, Hole A G2 and Hole A G3. Hole B and Hole C consist of one hole group. Among the hole groups in the +Z hole direction, the height of Hole B is higher than that of the other two hole groups of Hole A G1 and Hole C. There are 17 operations in total: 4 operations for Hole B, 4 operations for Hole C, and 9 operations for the operations of Hole A (three hole groups multiplied by three tools). The operation sequences are planned by following these three steps: (a) definition of the priorities for the sequence element (b) calculation of the score of each operation using the priorities, and
Tool number
Tool
Tool priority
#1 #2 #3 #4 #5 #6
C-drill Drill 18 Drill 50 Chamfer 40 Bore 30 Tap M20 × 2.5
1 2 3 5 4 6
Table 3 Hole group priority. Index
Hole group
Tool direction
Group priority
1 2 3 4 5
HoleA G1 HoleB G1 HoleC G1 HoleA G2 HoleA G3
(0, 0, 1) (0, 0, 1) (0, 0, 1) (0,-1, 0) (1, 0, 0)
1 1 1 2 3
(c) sorting of the operations by score in an increasing order. Using these steps, a feasible solution of the operation sequence is calculated. The details for each step are as follows: (1) Priorities for the sequencing The types of priority for the sequencing are listed below: • Nested cutting: If the hole-making process has completed from the first tool to the final tool, it is categorized to nested cutting and placed in the preceding position. • Tool number: In Table 2, a sample priority for tool number is defined. This element categorizes the operations according to the number of tools while operations need to minimize the number of tool changes. • Hole group (tool direction): The priority of hole group is illustrated in Table 3. The holes with a same direction tend to have an identical value and the planner selects the order of the machining directions. • Tool diameter: Tool priorities are determined in the ascending order of tool size as shown in Table 2. • Hole height: As shown in Fig. 6, holes of the Hole B G1 group are deeper than those of the Hole A G1 and HoleC G1 groups. Therefore, holes of the Hole B G1 group are set to the reference height. The height of these holes is set to zero point and the remaining groups are ordered by their heights. (2) Calculation of the score and determination of the sequence The operation sequence is influenced by the priority of the sequence element. If the priority of the sequence element is well defined, operations can be arranged to satisfy the requirements of the user. The score of individual operation is calculated as a function of the priority of the sequence element. The score of each operation is calculated as; Score = −En.c. × N 4 + Et.p. × N 3 + Eh.g. × N 2 + Et.d. × N 1 + Eh.h. × N 0
(1)
where En.c., the element of the nested cutting; Et.p., the element of the tool priority; Eh.g., the element of the hole group priority; Et.d., the element of the hole diameter; Eh.h., the element of the hole height; Mn.c., the max value of the nested cutting property; Mt.p. = the max value of the tool priority; Mh.g., the max value of the hole group priority; Mt.d., the max value of the hole diameter; Mh.h., the max value of the hole height; and N, max (Mn.c., Mt.p., Mh.g., Mt.d., Mh.h.).
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
119
Table 4 Examples of the operation sequence. ID
Hole group
Tool
Priority rule
Tool number
Hole group
Tool diameter
Hole height
3
2
1
0
(a) Case 1 1 10 14 4 7 2 11 15 5 8 3 6 9 12 13 16 17
HoleA G1 HoleB G1 HoleC G1 HoleA G2 HoleA G3 HoleA G1 HoleB G1 HoleC G1 HoleA G2 HoleA G3 HoleA G1 HoleA G2 HoleA G3 HoleB G1 HoleB G1 HoleC G1 HoleC G1
C-drill C-drill C-drill C-drill C-drill Drill 18 Drill 18 Drill 18 Drill 18 Drill 18 Drill 30 Drill 30 Drill 30 Bore 30 Tap 20 Chamf. 40 Tap 20
1 1 1 1 1 2 2 2 2 2 3 3 3 4 5 5 6
1 1 1 2 3 1 1 1 2 3 1 2 3 1 1 1 1
0 0 0 0 0 18 18 18 18 18 30 30 30 30 20 40 20
0 0 30 0 30 0 0 30 0 30 0 0 30 0 0 30 30
(b) Case 2 1 10 14 2 11 16 3 12 13 16 17 4 5 6 7 8 9
HoleA G1 HoleB G1 HoleC G1 HoleA G1 HoleB G1 HoleC G1 HoleA G1 HoleB G1 HoleB G1 HoleC G1 HoleC G1 HoleA G2 HoleA G2 HoleA G2 HoleA G3 HoleA G3 HoleA G3
C-drill C-drill C-drill Drill 18 Drill 18 Drill 18 Drill 18 Bore 30 Tap 20 Chamf. 40 Tap 20 C-drill Drill 18 Drill 30 C-drill Drill 18 Drill 30
1 1 1 1 1 1 1 1 1 1 1 2 2 2 3 3 3
1 1 1 2 2 2 3 4 5 5 6 1 2 3 1 2 3
0 0 0 18 18 18 30 30 20 40 20 0 18 30 0 18 30
0 0 30 0 0 30 0 0 0 30 30 0 0 0 30 30 30
(c) Case 3 1 10 2 11 3 12 13 14 15 16 17 4 5 6 7 8 9
HoleA G1 HoleB G1 HoleA G1 HoleB G1 HoleA G1 HoleB G1 HoleB G1 HoleC G1 HoleC G1 HoleC G1 HoleC G1 HoleA G2 HoleA G2 HoleA G2 HoleA G3 HoleA G3 HoleA G3
C-drill C-drill Drill 18 Drill 18 Drill 30 Bore 30 Tap 20 C-drill Drill 18 Chamf. 40 Tap 20 C-drill Drill 18 Drill 30 C-drill Drill 18 Drill 30
1 1 1 1 1 1 1 1 1 1 1 2 2 2 3 3 3
0 0 0 0 0 0 0 30 30 30 30 0 0 0 30 30 30
1 1 2 2 3 4 5 1 2 5 6 1 2 3 1 2 3
0 0 18 18 30 30 20 0 18 40 20 0 18 30 0 18 30
N is the weight value to avoid interferences and is the maximum among sequence elements. Generally, when N is 1000, there are no interferences. The feasible cases of the elements priority are: Case 1: classification by the number of tool changes, Case 2: classification by the tool direction, and Case 3: classification by the hole height of the same tool direction.
priority for tool number is higher than the priority for hole group. The score of ID = 1 (HoleA G1 Center Drill) for Case 1 in Table 4 is calculated as: 1, 001, 000, 000 = −0 × 10004 + 1 × 10003 + 1 × 10002 + 0 × 10001
(2)
The discussion for each case is as follows: Case 1. Classification by the number of tool changes The objective of Case 1 is to minimize the number of tool changes. The priorities (nested cutting, tool number, hole group, tool diameter, and hole height) are given as 4, 3, 2, 1, and 0. The
Table 4(a) shows the classified operations. Tool Number has the first order and Hole Group has the second order. Case 2: classification by the tool direction, and This case classifies the operations with the hole direction. The priorities are given as 4, 2, 3, 1 and 0. Hole group has a higher
120
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
Fig. 7. Multi-axis machine: (a) frame box of an engine block and (b) cylinder header.
order than tool number. This case can be useful when the cutting direction does not change. Case 3: classification by the hole height of the same tool direction. In this case, the priorities are given as 4, 1, 3, 0 and 2. To classify the holes with same heights and tool diameters, the priority elements are set as: hole group = 3 and hole heigh t = 2. The operations with zero height precede operations of the height of 30 as shown in Table 4(c). Nested cutting element has the highest priority. In the score function, the value of nested cutting element is negative. Accordingly, the operation checks whether nested cutting element is located at the first position of the operation list. The other values are all positive. If nested cutting elements for HoleA G1 operations have a value of 1 and if those for HoleA G3 have a value of 2, the operations of HoleA G3 are executed first. After that, the operations of HoleA G1 are executed as shown in Table 4(a). The proposed system allows a user to edit the priorities under the consideration of the tool sequence constraints defined in tool definition. In Case 1 of Table 4, the 6th operation (ID = 2, Drill 18 of Hole A-G1) is interchangeable with operations ranging from the 1st operation (ID = 1, C-drill of Hole A-G1) to the 11th operation (ID = 2, Drill 30 of Hole A-G1).
(1) Prismatic part A prismatic part of an engine block is huge in its size while the height can be as much as 4–5 m and length may be more than 10 m; moreover, holes may exist in deep and narrow places. An attachment enables access to these holes and to machine the holes in all directions except for the bottom direction. The hole machining method is similar to the conventional 2D drilling except for the tool axis changing. To support the tool and attachment changing, an additional tool path trajectory and attachment change codes are added. The additional tool path trajectory should be calculated for avoiding collisions between the moving objects (the attachment and tool) and the work piece. The attachment information of each machine is stored in the system database in advance. Thus, the number of the suitable attachments and the related parameters can be easily retrieved and inserted as shown in Fig. 7(a). (2) Cylindrical part
2.4.2. Tool path planning The tool path can be optimized by the traveling salesman problem (TSP) approach [5–8]. In other cases, holes should be categorized by the specific direction or the hole height. As shown in Table 4(a), the 1st and 2nd operations share the same tool number and height therefore these two operations are combined. The y-axis element of the distance vector of two points is higher weighted than the x-axis element of that in the TSP. In the case of many holes, the holes are pre-grouped by direction. Each group is considered as one hole in the TSP. In this way, the tool path can be planned in the x direction preferred rule. 2.4.3. Multi-axis machining To machine a part of a marine engine, multi-axis machining is essential. The shape of the part, as shown in Fig. 7, can be prismatic or cylindrical, and are described below.
Fig. 8. Interference checking.
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
121
Fig. 9. Main system flow for an engine block.
A cylindrical part is machined by a four-axis or five-axis machine with a turning table. The position and the orientation vector of a hole are calculated on the basis of a machine coordination system (MCS). If the holes resemble the spokes of a wheel, the MCS should be defined at the center of the circularpatterned holes as shown in Fig. 7(b) before generating NC data. Hence, the local MCS is defined at the center of the circularpatterned holes. The system generates an NC code based on the local MCS. 2.4.4. Check of error and interference A marine engine is heavy and expensive in nature therefore the number of errors must be minimized during the machining operations. When the tool direction and the tool number change, the
tool path and the cutting operations need to be validated. One of the challenges in tool path generation is to generate interferencefree tool paths. A cutter and stock interference takes place at the point-to-point and operation-to-operation links. As to the pointto-point linking, if there is an interference found between the stock and the cutter, the parameters of the engagement and the retraction should be recalculated by the following steps. Firstly, an interference-free height between the stock and the tool path is calculated, and then the offset height considering cutter and attachment movements is added to that height as shown in Fig. 8(a). As to the operation-to-operation linking, an additional path to escape collisions is calculated. Generally, passage point of path is calculated by the extreme value of the stock boundary box. Alternatively, it can be edited by user-defined position as shown in Fig. 8(b).
122
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123
Fig. 10. (a) Engine block and (b) operation list.
Fig. 11. (a) Cylinder header and (b) MCS and job name.
2.5. Operation Manager Operation Manager handles all of the different types of operations, including milling, turning, hole-making as well as the post-process information. This enables user to simulate, verify and check using various criteria. Operation Manager also allows operation editing. 3. Implementation result With the proposed concept, a hole-making system has been developed and Fig. 9 shows the main system flow with an engine block sample. The sample is selected for a case study, but the more complicated shapes of engine parts are commonly found in the industrial field. Fig. 9(b) is the main dialog, which can select and manage hole-types registered for machining beforehand. At each hole-type, the machining methods and rules are defined and hole-type instances are selected, as shown in Fig. 9(c). Then, the proper machining sequences are calculated (Fig. 9(d)) according to the cutting priority rules, and the results can be edited with easy, if additional conditions are changed. The final results are registered to the operation manager in Fig. 9(e). In the previous system, the machining sequences are defined manually in the operation manger step, thus it requires many hours to create and edit the machining data. For efficiency, however, the proposed system supports additional modules to handle previous machining data, as well as to secure the flexibility in machining sequence determination. Compared to the previous commercial system, the increased efficiency of the proposed system has been reported more than 60% in the actual industrial field. 3.1. Engine block Fig. 10(a) shows a machined engine block. The machined area was cut by a face-milling operation and all of the holes were machined using the proposed method. The operation sequence is categorized by the tool number, as shown in Fig. 10(b). 3.2. Cylinder header The cylinder header was machined by the four-axis and five-axis machines, as shown in Fig. 11(a). A cylindrical coordinate system
was applied to the circular-patterned holes and then the nested cutting was applied to the complex holes. In the operation list as shown in Fig. 11(b), MCS and its NC code (G54, G55, etc.) were added to each operation of the post-processing step. 4. Conclusions In this paper, a hole-making CAPP system is proposed to machine the parts of a marine engine. It consists of Hole Manager, Cutting Sequence Definition and Operation Manager. These components are derived from a conventional system. To satisfy the requirements, rules are defined for each module. For the operation sequence planning, three steps are proposed: (a) definition of sequence elements such as nested cutting, the tool number, the tool direction, the tool diameter, and the hole height, (b) score calculations for each operation, and (c) sorting of the operations by scores in an ascending order. By changing the priority of elements, the operation sequence can be adjusted. This idea is relatively simple but it is capable of yielding a significant efficiency along with a high flexibility. The proposed method was implemented in the commercial CAD/CAM system and utilized in industrial applications. Conflict of interest The authors have no conflict of interest. Acknowledgement This research was partly supported Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2011-0022044). References [1] Denkena B, Shpitalni M, Kowalski P, Molcho G, Zipori Y. Knowledge management in process planning. Annals of the CIRP 2007;56(1):175–80. [2] Park SC. Knowledge capturing methodology in process planning. ComputerAided Design 2003;35:1109–17. [3] Wong TN, Chan LCF, Lau HCW. Machining process sequencing with fuzzy expert system and genetic algorithms. Engineering with Computers 2003;19:191–202.
C.-S. Lee et al. / Journal of Manufacturing Systems 32 (2013) 114–123 [4] Yongtao H, Jingying M. A knowledge-based auto-reasoning methodology in hole-machining process planning. Computers in Industry 2006;57(4):297–304. [5] Kiritsis D, Neuendorf KP, Xirouchakis P. Petri-net techniques for process planning cost estimation. Annals of the CIRP 1998;47(1):427–30. [6] Lee DH, Kiritsis D, Xirouchakis P. Branch and fathoming algorithms for operation sequencing in process planning. Internal Journal of Production Research 2001;39(1):1649–69. [7] Lee DH, Kiritsis P, Xirouchakis P. Simulated annealing algorithms for operation sequencing in nonlinear process planning. Journal of the Korean Institute of Industrial Engineers 2001;27(3):315–27. [8] Lee CS, Kim SS, Choi JS. Operation sequence and tool selection in flexible manufacturing systems under dynamic tool allocation. Computers and Industrial Engineering 2003;45:61–73. [9] Etienne A, Dantan JY, Siadat A, Martin P. An improved approach for automatic process plan generation of complex borings. Computers in Industry 2006;57(7):663–75. [10] Kalahan F, Liang M. Hybrid genetic algorithm and simulated annealing approach for the optimization of process plans for prismatic parts. The International Journal of Product and Research 2002;40(8):899–922. [11] Ghaiebi H, Solimanpur M. An ant algorithm for optimization of hole-making operations. Computers and Industrial Engineering 2007;52(2):308–19. [12] Kalahan F, Liang M. A tabu search approach to optimization of drilling operation. Computers and Industrial Engineering 1996;31(1/2):371–4. [13] Onwubolu GC, Clerk M. Optimal path for automated drilling operations by a new heuristic approach using particle swarm optimization. International Journal of Production Research 2004;42(3):473–91. [14] Li J, Gao S, Liu Y. Solid-based CAPP for surface micromachined MEMS devices. Computer-Aided Design 2007;39(3):190–201. [15] Smith S, Zamudia C. Activity model and computer aided system for defining sheet metal process planning. Journal of Materials Processing Technology 2006;173:213–22.
123
[16] Zhou X, Qiu Y, Hua G, Wang H, Ruan X. A feasible approach to the integration of CAD and CAPP. Computer-Aided Design 2007;39(4):324–38. [17] Park HS. A knowledge-based system for assembly sequence planning. International Journal of Precision Engineering and Manufacturing 2005;1(2): 35–42. [18] Park HS, Choi HW. Disassembly process planning of end-of-life car. International Journal of Precision Engineering and Manufacturing 2005;6(1): 42–50. [19] Denkena B, et al. Knowledge management in process planning. CIRP Annals – Manufacturing Technology 2007;56(1):175–80. [20] Halevi G, Wang K. Knowledge based manufacturing system (KBMS). Journal of Intelligent Manufacturing 2007;18(4):467–74. [21] Wang L, Feng HY, Cai N. Architecture design for distributed process planning. Journal of Manufacturing Systems 2003;22(2):99–115. [22] Liu Z, Wang L. Sequencing of interacting prismatic machining features for process planning. Computers in Industry 2007;58(4):295–303. [23] Wang L, Cai N, Feng HY, Liu Z. Enriched machining feature based reasoning for generic machining process sequencing. International Journal of Production Research 2006;44(8):1479–501. [24] Wang L, Adamson G, Holm M, Moore P. A review of function blocks for process planning and control of manufacturing equipment. Journal of Manufacturing Systems 2012;31(3):269–79. [25] Xu X, Wang L, Newman ST. Computer-aided process planning – a critical review of recent developments and future trends. International Journal of Computer Integrated Manufacturing 2011;24(1):1–31. [26] Han J, Pratt M, Regli WC. Manufacturing feature recognition from solid models: a status report. IEEE Transactions on Robotics and Automation 2000;19(6):782–96. [27] Joshi N, Dutta D. Feature simplification techniques for freeform surface models. Journal of Computing and Information Science in Engineering 2003;3(9):177–86.