|
|
|
|||||||||
|
|||||||||
|
|||||||||
By Andreas Roderburg, Fritz Klocke and Christoph Zeppenfeld
Current developments of hybrid production processes or production systems have exceeded performance in manufacturing. What most of these developments have in common is that they have been found to be intuitive. In the past, the development of new hybrid production systems have led to large amounts of operative planning. The challenge is to develop a systematic and scientific approach for aggregating, describing, explaining and combining single processes. This paper introduces a systematic approach of the design methodology for developing hybrid production processes. The Theory of Inventive Problem Solving (TRIZ) tools are used in terms of identifying hybrid process solutions as part of an innovation process. They are also shown concerning the specific requirements of manufacturing process development.
The manufacturing of innovative products often fails due to technological limitations of today's production systems and processes. The hybridization of manufacturing processes can help to break the limits of today's production systems and to realize innovative products. Developments of new hybrid manufacturing technologies, thus, arouse high expectations.
Previous developments of hybrid machining processes are mostly based on intuitive or random solutions. A methodology for the systematic development of hybrid machining technologies does not exist. But as is known from other scientific disciplines, a big part of solutions can not be found effectively without a methodological approach.1,2,3,4 It can be assumed, therefore, that up to now the full potential of hybrid technologies has not been exploited.
Within hybrid production processes different forms of energy or forms of energy caused in different ways are used at the same time at the same zone of impact. Hybrid production processes are also defined as the combination of effects that are conventionally caused by separated processes in one single process at the same time.5
The realization of hybrid production processes is based on two approaches. First, through the integration of specific forms of energy in the work area of the process; where the machinability of the workpiece material is improved. Or second, through the process mechanisms where characteristics are improved.
For the realization of these approaches, one or more supplementary forms of energy have to be coupled into the existing process. For this purpose several different forms of energy combinations are possible consisting of thermal, mechanical or chemical energy, although the technological advantage and the possibilities of realization must be considered.6
For the majority of companies in the production industry – especially in high wage countries – innovation is one of the most important factors of success. In regards to this matter it is not only the product idea that contributes to a higher success, but also the capability of optimal production processes. (In this context "optimal" means most efficient, flexible and stable at the same time.) Broad manufacturing process know-how is necessary to run a technology near its limits of performance. In most cases using current manufacturing technologies (by means of conventional optimization) does not lead to a leap in performance. Hybridization of conventional technologies is a way of reaching new technology capabilities in production systems. Figure 1 illustrates how hybrid processes can improve a production in terms of process chain shortening or the realization of new product qualities.
| |||||
The described technology integrations of the process chain can be shortened due to the substitution of several processes by a single hybrid process. This can lead to a reduction of planning operations that are necessary to synchronize several processes in a process chain, even though a hybrid process may be more complex and may cause more requirements concerning process stability. A second aim of developing hybrid processes is to realize new product qualities. For example, by hybridization a conventional process can be enabled to a machine for new materials, which according to the state of the art, could not yet be machined by the conventional process. Based on these new production capabilities innovative products could be realized, which could not be produced (economically), due to the technological limitations of today's production technologies.
The research and development of production technologies often focuses on single production processes concerning specific applications. This research is specialized and vital for understanding the mechanisms of known production processes. The knowledge about the mechanisms of single technologies enables these production processes to be adapted and optimized for a wide field of specific product requirements. According to the S-curve theory the possibility of process improvement decreases after a specific time when considering single processes. In this state the technology is optimized near its limits. In the case of conventional, widely investigated manufacturing technologies, production research often focuses on adapting the known technologies for new applications such as the machining of new workpiece materials. On the other hand, more capable and efficient manufacturing technologies have to be developed. Assume that the possibilities of hybrid production technologies is not utilized. The development of new inventive production technologies has to consider and combine the mechanisms of different single processes. For this purpose the integration of profound process knowledge from different production technology disciplines is necessary. In order to provide the broad field of necessary knowledge it has to be compressed in an abstract way to get new ideas, which is the first step of inventive solutions. The aim of the described scientific work is to develop a systematic tool for:
To find inventive (hybrid) process solutions, different steps have to be proceeded. Figure 2 shows a general structure of the methodology and its similarities to the general TRIZ algorithm.
| |||||
The presented approach is based on the assumption that a new process solution can be developed only if the limits of a technology are known with their reasons and can be described. The limits of technologies can be characterized in terms of problems, while a problem is defined as a contradiction. Thus, in the first steps of the methodology the problems of today's technologies must be analyzed systematically. The Theory of Inventive Problem Solving provides several tools for the purpose of analyzing problems. In accordance with TRIZ the technology limits are described in terms of standard problems through abstract physical or technical contradictions. By means of TRIZ, standard solutions can be identified for standard problems (by the contradiction matrix, separation principles or standard solutions for substance field models).7,8,9,10
Based on an aggregation of solutions and mechanisms of known production processes – as well as single and hybrid processes – analogies can be provided to support the search for a specific solution. For example, if a local heating of the workpiece surface layer is required then the use of laser beams is a specific solution. This is known from laser-supported turning or laser-supported incremental sheet metal forming.
Technology capabilities can be described by several criteria composed of output quantities or output characteristics, which are related to the resulting workpiece, the tool wear or the productivity (see Figure 3). While the tool wear and the productivity are primarily economical evaluation criteria, the workpiece results are technological criteria.
| |||||
The single manufacturing process can be described in terms of a black box by its input and output parameters, as shown in Figure 4. This kind of description model reduces the complexity of a manufacturing process. In manufacturing research the correlations between input and output parameters of a specific process are often described by empiric-analytical models, which are based on experimental investigations. The results of these investigations are used for the purpose of process chain generation. In a process chain the workpiece state output of one single process is the workpiece state input of the following process until the desired workpiece condition is reached.
| |||||
The description of the workpiece state before and after the manufacturing process classifies the workpiece characteristics into macro geometry, surface topography, surface layer and bulk features. The first procedure introduced methodology where a technology is described concerning the results of the workpiece surface topography and the influence on the workpiece material characteristics. For this purpose a workpiece element is defined. The difference between the states of the workpiece element (before and after a manufacturing process) describes the manipulation of the workpiece by the process. For the use of process chain generation the considered technology is sufficiently described by the change of the workpiece element characteristics. Figure 5 shows an example of a workpiece influenced by a grinding process. The surface layer consists of a re-hardening zone and an annealing zone. This manipulation is also known as "grinding burn."
| |||||
In general, technological limits are derived from a matching of required product results and technology capabilities. The realization of some requirements is often directly opposed. The better the first criterion the worse the other one gets. In most technologies the characteristic conflicting output criteria process includes examples of accuracy and productivity, or tool wear and productivity. Figure 6 illustrates this in a conventional grinding process.
The assumed object is designed to optimize a conventional grinding process. A capability profile of the grinding process shows the following relation tendencies among some important output parameters of a grinding process:
| |||||
If the surface roughness of the grinding process should be improved then the infeed speed can be reduced. This causes longer process times and leads to lower productivity. If the productivity should be constant, the surface roughness can be improved by a higher cutting speed or by the use of another grinding wheel specification that features a lower grain size. Both lead to a higher risk of grinding burn.
As shown in Figure 6, undesirable process outputs and generally conflicting outputs are aggregated for many different manufacturing technologies. This is the basic model for a broad description of limits in production technologies.
While the described first step of the methodology features a phenomenological/descriptive view, in the following step the interrelationships of the process are considered by means of cause-and-effect. The change of view from the phenomenological to an interpretational view is also expressed by the terms that are used within the process models. Within the description level the terms of input and output mainly have been used. Within the interpretation level technology outputs are understood as effects at the end of a cause-and-effect chain. The production process is therefore no longer described as a black box, but it is characterized by a cause-and-effect chain as shown in Figure 7.
Compared to the descriptive process model in Figure 4 the process is divided into two sub-models. The first sub-model "Interaction" is technology specific. It represents the mechanisms of interaction among the production system elements such as the interactions between tool and workpiece. Within these interactions energies are transmuted, which means that the kinematic energy of a moving tool or workpiece is transmuted into work of friction and deformation, leading to mechanical and thermal energy as an effect of the interaction. The effected energies are defined as process characteristics. As a function of place and time these process characteristics describe the production process in a technology unspecific way without the need of profound information about the process input parameters. The unspecific process characteristics of a technology are vital for a common semantic for the purpose of comparing and combining different technologies.
| |||||
The opposite of the sub-model Interaction is the second sub-model: "Impact" and it is technology unspecific. It represents the impact of the process characteristics on the process results such as the workpiece features and the tool wear. Within this sub-model the process characteristics are the root causes of the results. Note: Figure 7 shows a reduced model in order to clarify the idea of the model. In the case of applying to a specific technology problem, it can be adapted or added to by further cause and effect elements, where the influence of specific machine specification supplies such as lubricants and external disturbances is not considered. The feedback of tool wear also is neglected in this model because in this case it is presumed that the principles of physical mechanisms do not change while tool wear appears.
| |||||
Today's manufacturing technology research and development often focuses on the optimization of single processes for specific applications by means of changing the tool specification and process setting parameters. This kind of process improvement is represented by the conventional optimization approaches as shown in the upper part of Figure 9. The need for radical changes of technologies cannot be achieved by these approaches.
Though the mechanisms of single technologies today have been investigated and modeled in detail by experts, interdisciplinary solutions are rarely found. The reasons for this include the limited individual knowledge concerning other disciplines or technologies as well as psychological barriers (psychological inertia vector). A link of different technology models is vital in order to increase the probability of interdisciplinary solutions in production technology research and development. The technology unspecific process characteristics are used in this approach as a common semantic and interface for different technology models.
After applying and executing all conventional optimization approaches, the lower part of Figure 9 helps illustrate another way for achieving a radical improvement in terms of clearly enhancing the technical limits of a known process.
| |||||
Figure 9: Assuming the possibilities of conventional optimization are nearly exploited, the process can only be improved by changing the composition of process characteristics by external (energy) sources. This integration is based on an interaction of the added technology and the workpiece or the conventional process as well.
Another possibility of finding a hybrid solution is to analyze the resources of process chains in terms of cause-and-effect relations. Based on this, a process chain may be improved by integrating single processes or the impacts of several processes in one technology.
The described design level of the methodology provides a way to find a hybrid production technology solution. The general structure of this approach is based on the principles of TRIZ. Several TRIZ tools have to be integrated to the introduced procedure in order to provide an effective applicable way that can extend the capabilities of technology development in production research.
The basic correspondences between TRIZ and the steps of hybrid production process design were shown in Figure 2. The parallels between TRIZ and the introduced production process design methodology show a high potential for the use of TRIZ within this context. The main parallels are shown in the following interdisciplinary solutions.7,8,9,10
For the purpose of finding hybrid solutions a combination of different technology models is necessary. Due to the psychological inertia vector (PIV) the non-methodological problem solving process mostly focuses on a small field of possible solutions, which is limited to the technological knowledge of an individual person. An interdisciplinary solution is often outside the focus of solution that is determined by the PIV. Such an inventive solution has to be found systematically with the support of several TRIZ tools.
Both TRIZ and the hybrid production process design methodology aim at the "ideal" solution instead of accepting early compromises. At least two conflicting output criteria of the manufacturing system have to be improved at the same time in order to achieve a technology improvement with higher ideality compared to conventional optimization.
The analogy approach is based on the assumption that most problems already have been solved in other disciplines or in another context. The new solutions, therefore, can often be found based on analogue solution principles that have been realized in other technical areas. This implies an aggregation of a known solution and an extracting of solution principles. In the field of manufacturing technologies, existing technology database systems can help to analyze existing solutions in order to find solutions for unsolved problem. In terms of an efficient production process design methodology technology is only limited knowledge that can be given and must be provided with purpose.
Known specific solutions can be used to identify standard solution strategies, which can be used to find a solution for another specific problem. The linking of solution and problem is enabled by an abstract modeling of the specific problem. This origin idea of TRIZ is introduced through the modeling of single processes. The problem of technology is described in an abstract manner by process characteristics based on their physical mechanisms. As a next step the solutions are generated on this abstraction level in the previous section titled: Design Level: Breaking the Limits.
In literature TRIZ is described as a methodology, consisting of different method tools. The table titled TRIZ Tools Overview summarizes several TRIZ tools as they are used for problem solving in a broad field of applications.
| TRIZ Tools Overview12 | |||
| Systematic | Knowledge | Analogy | Vision |
| Innovation checklist | Physical effects | 40 inventive principles | S-curve |
| Resource checklist | Internet search | Contradiction matrix | Laws of evolution |
| Ideality | Patent search | 4 separation principles | |
| Operator MTC | Substance-field model | ||
| Miniature dwarfs | |||
| Problem formulating | |||
With the integration of TRIZ and the introduced methodology of production process design, the applicability of several TRIZ tools is evaluated by means of known hybrid production processes. The physical mechanisms of single processes are considered and process solutions are developed by the help of TRIZ tools.
The application of the described methodology in combination with several TRIZ tools is exemplified by the technology evolution from grinding to grind-hardening as a new hybrid application of the grinding process.
Grinding is generally used as a finishing process of hard materials due to its high dimensional and shape accuracy, although grinding can be used for high stock removal rates as well.
In grinding of hardened steel, undesirable effects can appear near to the ground surface such as crack formation, workpiece material structure transformation, modification of hardness progression and residual tensile stress. As mentioned in the section titled: Description Level: Identification of Limits, the workpiece material structure transmutations (also called grinding burn) can be divided into a white etching re-hardening zone and an annealing zone, see Figure 5. The annealing zone features a lower hardness than the bulk material. The re-hardening layer is harder and more brittle compared to the bulk material. The general conflicting output criteria of a grinding process concerning grinding burn is shown in Figure 6.
The progression of temperature over time is the main influence factor of dimension and modality of the surface layer. The temperature progression can be characterized by the maximum temperature and time period of thermal impact.13 Figure 10 shows the cause-and-effect relationship of the grinding process regarding the undesirable effect of grinding burn. This occurs if the productivity is raised to a critical point while the surface roughness and tool wear is constant.
| |||||
It is assumed that all conventional optimization approaches, according to Figure 9, are exhausted. The root causes in terms of setting parameters, therefore, are not considered in Figure 10. As a first step it is also sufficient not to regard the interaction of tool and workpiece. Instead the first object of process improvement is the level of process characteristics, which are assumed to result from the process interactions. In contrast to the other system and process setting parameters, the initial workpiece material features have turned out to be important information, which has to be considered on every abstraction level of the process model. It is a crucial fact that the effects of interaction as well as the effects of process impact such as material structure transmutations depend largely on the initial material state.
In order to find a solution for the described problem in terms of TRIZ, the first step is to define the problem in the form of a contradiction. In Figure 11 the reduced substance field model of the grinding process depicts the problem.
Between the grinding wheel and the workpiece a mechanical field is generated effecting the desirable chip formation and the material removal. A thermal field is generated as a second field that causes the undesirable effect of thermal workpiece surface damages.
| |||||
Due to physical laws in conventional grinding the amounts of thermal and mechanical energy are correlated. Both are affected by the total amount of process energy. The use of standard solutions has brought out some known solutions such as using cooling lubricants. Cooling lubricants can improve the ratio between the mechanical and the thermal fields. Up to now no other solution was found to resolve this relation. The problem can then be described by a physical contradiction:
A possible solution is to apply the principle of separation in time. While in a rough machining process the productivity is high, a damaged surface layer can be compensated afterward by a finishing process. This solution is well known and probably the most applied one in the industry, but it is not a hybrid process solution. To find another solution for this particular problem is a big challenge for future research.
Another approach is to change the view of examination to the superior system of process chains. As long as there is no specific application to be considered, it is appropriate to describe a process chain. This is commonly linked to grinding technologies. Grinding is mostly used for hard machining and the grinding process often follows a heat treatment process. Here the heat treatment is applied in order to harden the surface layer. The considered process chain consists of soft machining, followed by heat treatment (induction hardening or laser hardening) and then by grinding for hard machining. The heat treatment processes cannot easily be integrated into the production line. For this reason the heat treatment process involves some more processes due to transport and cleaning operations that are time and cost consuming, see Figure 12.
| |||||
Through an expansion of the cause-and-effect description to the process chain, the TRIZ tools innovation checklist (ICL) and resource checklist (RCL) can be used. In terms of this process design methodology (not the technologies) their abstract cause-and-effect relationships are considered as the resources of a process chain. Within the considered process chain an accumulation and comparison of cause-and-effect models of all participating technologies shows a common cause-and-effect relationship shared by the grinding and hardening process, as shown in Figure 13.
| |||||
Concerning the thermal aspects of the processes the same process characteristics determine the thermal impact onto the workpiece surface layer. While the effect of material structure transformation in grinding leads to undesired grinding burn the material structure transformation in hardening causes the desired process result. The crucial difference between the two technologies is that the initial state of workpiece material of a grinding process is a hardened material, but in hardening the material is unhardened beforehand. Using the principles of ICL/RCL it can be concluded that grinding of hardened material can lead to undesired grinding burn, but in grinding of unhardened materials it may be possible to use the heat progression for the purpose of surface hardening.14 Figure 14 shows the principle of grind-hardening.
| |||||
As primary experimental investigations have shown, the heat generated in grinding can be utilized for thermal induced surface hardening effect up to mm-scale. They have shown that the metallurgical effects in grind-hardening are comparable to those of conventional surface hardening technologies such as laser and induction hardening.14,15,16,17,18
By the use of grind-hardening a separate hardening process can be substituted and the involved time and cost consuming transport and cleaning processes become unnecessary, see Figure 15.
| |||||
The result: A new application of grinding technology was found in using the thermal source of a grinding process for surface material hardening.
This paper introduced a systematic approach of a design methodology for developing hybrid production processes that can clearly enhance the capability limits of today's production technology systems. According to the principles of TRIZ the introduced approach consists of three steps.
These abstract solutions are the origin for the integration of different technologies and for new approaches in terms of breaking the limitations of single technologies by radical changes. The Theory of Inventive Problem Solving provides several method tools that support the search for inventive (hybrid) production process solutions and that have to be integrated. As illustrated in the case of grind-hardening the introduced systematic approach supports the technology development in terms of generating new ideas and solutions for enhancing the limits of today's production technologies.
The authors would like to thank the German Research Foundation and RWTHAACHEN University DFG for the support of the depicted research within the Cluster of Excellence "Integrative Production Technology for High-Wage Countries."19
This paper was originally presented at the European TRIZ Association's TRIZ Future 2008 meeting in Enschede, NL.
Andreas Roderburg is part of the Laboratory for Machine Tools and Production Engineering at RWTHAACHEN University. Contact Andreas Roderburg at A.Roderburg (at) wzl.rwth-aachen.de.
Professor Fritz Klocke obtained a PhD in Production Engineering from the Technical University of Berlin in 1982. In 1995 he was appointed as Professor for Manufacturing Technology at WZL of the RWTH Aachen University and Director of the Fraunhofer Institute for Production Engineering. He was the President of the International Academy for Production Engineering (CIRP) for the period 2007/2008.
Dr. Christoph Zeppenfeld obtained a PhD in Production Engineering from RWTH Aachen University in 2005. Currently he is the Chief Technical Officer at the Chair for Manufacturing Technology at the WZL.
