Suitability analysis of using industrial robots in manufacturing.
Suitability analysis is the process and procedure used to establish a system that meets the needs of users. Suitability of robotization is a basic question for managers who are planning changes in the company. For producing goods, companies have to perform different processes and industrial tasks. There are certain aspects why industrial robots are used for those processes. Theoretically, the basic aspects concern humans, productivity and quality [1-3].
The widest areas using industrial robots are: welding, machine tool servicing, assembling, painting, loading-unloading, packaging, palletizing, and medical applications. Welding ranks among the most important joining processes and has special features for the industrial robots, such as programming task sequences, free definition and parameterization of robot positions/orientations, high repeatability and positioning accuracy of moving paths, high speed of end effector, minimum six degrees of freedom (6DOF), variable payloads depending on the welded products (2-150 kg), advanced programmable logic controllers (PLCs). The vision, strategy and action plans for implementation of robots are described in  and statistics about using robots in industry can be found in .
Literature review and efficiency analysis of IR cells are important basis for gathering information. To estimate the suitability of using industrial robots in different application areas, it is necessary to analyse the applications of the robot in the industry and to solve a decision-making task.
Suitability analysis belongs to the tasks of dual approach. From one hand, it is an application area for the efficient use of industrial robots and on the other hand it supports decision making methods. The decision maker (expert) must have an excellent understanding about the application area and should be familiar with the factors influencing the effective use of industrial robots. For this purpose, a robot-based manufacturing cell performance evaluation conceptual model was developed, which is based on a recursive decision-making procedure . In the model shown in Fig. 1, there are four groups of parameters: product features, robot cell features, elements of evaluation and general output description. The first two groups are the parameters of the design level (parameters of the product portfolio and their manufacturing processes) and the last two groups are the execution outcomes (different KPIs, that measure, and critical success factors--cost factors, level of achieving the general objectives, dynamics of effectiveness, employee competencies, etc.). These interactions reflect the suitability of using the real IR cell in the company.
The main concern in the suitability analysis process is to find the best solution, according to the set of criteria using the method, which allows the most realistic input (importance) of each criterion. There are different possibilities to influence the roles of criteria: the equal weight (EQW) heuristic , the weighted additive (WADD) rule . However, the main risk is overestimating some of the criteria or not paying enough attention to others. Therefore, artificial intelligence (AI) methods may be used .
Analysing industrial robot's applications for welding, there were developed three general groups of parameters, shown in Table 1. Those listed parameters have the greatest influence to the suitability of using welding robots in the company.
Having knowledge about the welding process and parameters make the welding process more efficient. It is possible to find the tools for suitability analysis. To solve the engineering task, such as making a decision about the suitability of welding robots in the company, multiple criteria should be used (see Table 1). Each decision corresponds to a variable, relation or predicate, whose possible values are listed among the condition alternatives. Each action is a procedure or operation to perform, and the entries specify whether or in what order the actions should be performed for the set of condition alternatives the entry corresponds to.
3. ANALYTIC HIERARCHY PROCESS
Solving the engineering decision making problem of IR cell suitability, which has multiple criteria and alternatives, is a difficult task. One of the techniques for solving multi-criteria decision making (MCDM) problems is analytic hierarchy process (AHP). AHP was proposed by Saaty  and developed further by [10,11]. Those methods use fundamental scale of relative importance to construct a pairwise comparison matrix of attributes. Likewise, consistent weight of attributes is determined, which help evaluating composite performance score of alternatives. The alternatives are then ranked according to their composite performance score. Several steps and principles should be considered and understood for constructing a MCDM problem solving tool. The steps are the following:
(1) Developing the hierarchy criteria model for decision;
(2) Deriving priorities by pair-wise comparison for the criteria. Pair-wise comparison scales are shown in Table 2;
(3) Determining local priorities for alternatives;
(4) Calculating and adjusting the consistency:
(a) Multiplying the matrix of judgements by the eigenvector, obtaining a new vector (A[omega]);
(b) Dividing each component of a new vector of A[omega] by the corresponding eigenvector element;
(c) The mean value from the point b is the estimated for [[lambda].sub.max];
(d) Calculating the consistency index (CI) by:
CI = ([[lambda].sub.max] - n) / n; (1)
(e) Calculating the consistency ratio (CR) by:
CR = CI / RI; (2)
(f) Checking the consistency of the hierarchy. CR should be below or equal to 10%;
(5) Populating the judgement matrix with input data: quantitative data, such as product payback period total investment etc. are normalized by using Eq. (3). Dimensions, mass etc., are normalized by using Eq. (4). Qualitative data, such as complexity of operations, manufactured parts precision, experience and competencies of engineering stuff and workers, etc., are graded by the scale of 1-5 and normalized.
Z = 1 - [[x - [x.sub.min]]/[[x.sub.max] - [x.sub.min]]], (3)
Z = [[x - [x.sub.min]]/[[x.sub.max] - [x.sub.min]]]; (4)
(6) Making the final decision.
4. CASE STUDY
Production cells were investigated in twenty SMEs with the number of employees ranging from 20 to 150. They produced different parts for agricultural and forestry machines, small tractors, high speed trains, lifts' components, wind generator rotors and other sheet metal products. The information was acquired by interviewing companies' management, engineering staff and data extracted from the enterprise resource planning (ERP) software. Data gained from the interviews and ERP system contained both, quantitative and qualitative data. From the collected data, the information about the robot welding production units was the only one used for the following suitability analysis.
Three performances of production units were stated as a benchmark for the suitability analysis. Production cells, shown in Table 3, were chosen by their excellent KPIs' outcomes. KPIs were selected according to the performance evaluation model  and they are as follows: discounted payback period (DPP), cell utilization (CU) and overall equipment effectiveness (OEE).
5. CRITERIA AND SUB-CRITERIA
Regarding this study, the main task of multi criteria decision analysis (MADM) is to estimate the suitability index and it is based on the following criteria and sub-criteria (see Fig. 2):
* Production unit (PU):
** cost (C): total investment (C1), cost of utilities (C2), running costs (C3);
** maintenance (M): maintenance cost (M1), emergency maintenance cost (M2);
** level (L): use of CAD/CAM (L1), automated storage (L2), machine vision (L3).
* Product (P):
** physical properties (PP): complexity of parts (T1), parts manufacturing precision (T4), mass (T6);
** productivity (PR): product families (T2), patch size (T3), patch repeatability (T9), overall welding ratio (TE3), average cycle time (TE9), average setup time (PR2), quality assurance (E2).
* Company environment (CE):
** workforce (WF): workstation fulfillment (E1), workers salary (E6), production engineer's involvement (E8), shifts (W2), durations of shifts (W3);
** performance indicators (PI): increment of productivity (E4), increment of on time delivery OTD (E9), increment overall equipment effectiveness OEE (E10), payback period (K1);
** experiences (E): experiences with MIG/MAG, TIG (TE1), competencies in welding technology (TE2), experiences with robotization (TE4), experiences with jigs and fixtures (TE7), workstation organization level (TE8), overall automatization level (TE10).
6. PERFORMANCE SCORES
The assessments obtained from the decision makers are made by pairwise comparisons. Performance scores and consistency ratio are calculated and given in Tables 4 and 5.
7. DECISION MATRIX
The normalized inputs are multiplied by their corresponding performance scores and the local and global scores are summed up. Results are shown in Table 6.
8. DISCUSSION AND CONCLUSIONS
In this study an AHP based suitability analysis for robot integrated production cells was developed. Twenty production cells in different industries and at different levels were investigated. Based on the literature, review input parameters were selected, criteria set up and hierarchy of the problem were developed. To ensure the objectivity of experts' pairwise comparisons of the responses of criteria, consistency ratio was calculated and controlled. For testing the developed tool, a case study approach was used. Three welding cells were selected based on their excellent KPIs' outcomes and set as a benchmark for suitability analysis. The highest overall suitability score was obtained in case of No. 3 with index of 0.17. The extremely high score was received in both, product and company environment categories, i.e. 0.849 and 0.810, respectively. The suitability analyses confirmed an excellent choice of product to be produced in a well-organized cell and automated company environment. For decision of suitability, four categories were proposed in Table 7, based on suitability criteria for robot welding, shown in Table 1.
For more precise results, it is possible to simulate the planned robot cell and to calculate the break-even point. Having enough competence in all these areas is quite sophisticated. Therefore, the tool which gives the possibilities to estimate the suitability of using industrial robots for the automation of a certain manufacturing process, is important in the early stage of planning. For future work, more robot integrated processes like machine tending, palletizing, etc. can be added to the tool.
We thank our colleagues from Taltech, IMECC and TTK UAS who provided insight and expertise that greatly assisted the research. The publication costs of this article were covered by the Estonian Academy of Sciences.
[1.] Miller, R. K. Industrial Robot Handbook. Springer, New York, 1989.
[2.] Nof, S. Y. Handbook of Industrial Robotics, Second Edition. John Wiley & Sons, New York, 1999.
[3.] Koren, Y. Robotics for Engineers. McGraw-Hill Book Company, New York, 1985.
[4.] New Robot Strategy. Japan's Robot Strategy. Vision, Strategy, Action Plan. https://www.meti.go.jp/english/press/2015/pdf/0123_01b.pdf
[5.] World Robotics Report 2016. International Federation of Robotics. https://ifr.org/ifr-press-releases/news/world-robotics-report-2016
[6.] Kangru, T., Riives, J., Otto, T., Pohlak, M., and Mahmood, K. Intelligent Decision Making Approach for Performance Evaluation of a Robot-Based Manufacturing Cell. In Proceedings of the International Mechanical Engineering Congress and Exposition, November 9-15, 2018, Pittsburgh, Pennsylvania, USA. https://asmedigitalcollection.asme.org/IMECE/proceedings-abstract/IMECE2018/52019/V002T02A092/276341
[7.] Pfeiffer, J. Fundamentals on Decision-Making Behavior. In Interactive Decision Aids in E-Commerce. Physica-Verlag HD, Berlin, Heidelberg, 2011, 15-45. https://link.springer.com/chapter/10.1007%2F978-3-7908-2769-9_2
[8.] Zhang, L., et al. Using Neural Network to Evaluate Construction Land Use Suitability. In Proceedings of the 2010 Second International Workshop on Education Technology and Computer Science, March 6-7, 2010, Wuhan, China. https://ieeexplore.ieee.org/document/5459823
[9.] Saaty, T. L. The Analytic Hierarchy Process. McGraw-Hill, New York, 1980.
[10.] Kaganski, S., Majak, J., and Karjust, K. Fuzzy AHP as a tool for prioritization of key performance indicators. Procedia CIRP, 2018, 72, 1227-1232.
[11.] Paavel, M., Karjust, K., and Majak, J. Development of a product lifecycle management model based on the fuzzy analytic hierarchy process. Proc. Est. Acad. Sci., 2017, 66 (3), 279-286.
Toostusrobotite kasutatavuse sobivusanaluus
Tavo Kangru, Juri Riives, Kashif Mahmood ja Tauno Otto
Tootmisettevottes on maarava tahtsusega enne robottootmisuksuse loomist labi viia toostusrobotite sobivusanaluus. Selle valjatootamiseks tehti Eesti vaikese ja keskmise suurusega ettevotete hulgas uuring, maaramaks toostusrobotite kasutust. Kogutud andmete pohjal viidi labi spetsiifilisem tootmisuksuste uuring, millega hinnati tootmisrakkude projekteerimisel pustitatud eesmarkide saavutamist. Tulemusena loodi toostusrobotite sobivuse hindamise meetod koos kriteeriumide ja tulemuslikkuse votmenaitajate kogumiga. Hindamismeetod voimaldab hinnata rakendatud tootmisuksuse kasumlikkust ja saada taiendavat majanduslik-tehnilist teavet tulevaste robottootmisuksuste arendamiseks.
Tavo Kangru (a*), Juri Riives (b), Kashif Mahmood (c), and Tauno Otto (c)
(a) Institute of Technology, TTK University of Applied Sciences, Parnu mnt 62, 10135 Tallinn, Estonia
(b) Innovative Manufacturing Engineering Systems Competence Centre, Teaduspargi 8, 12618 Tallinn, Estonia
(c) Department of Mechanical and Industrial Engineering, Tallinn University of Technology, Ehitajate tee 5, 19086 Tallinn, Estonia
Received 12 April 2019, accepted 9 May 2019, available online 28 October 2019
(*) Corresponding author, firstname.lastname@example.org
Table 1. Suitability criteria for robot welding Product view Technology view 1. The products are complicated from 1. Experiences in MIG/MAG the technological point of view and TIG welding 2. The products can be classified into 2. Competences in welding product families technologies 3. The products are produced in 3. Welding processes have repeatable batches great importance in the company's production processes 4. The products are of high quality 4. There are already 5. It is necessary to use welding fixtures experiences with robot welding 5. It is necessary to increase the productivity of welding processes Objectives' view 1. To shorten the throughput time 2. To increase the productivity in the workplace 3. To increase the product quality 4. To increase the precision of delivery 5. To reduce the product cost Table 2. Pair-wise comparison scale assessment Importance Description 1 Equal Importance 3 Moderate importance 5 Strong importance 7 Very strong importance 9 Extreme importance 2, 4, 6, 8 Values between two adjacent values should be in considerations Inverse If activity (i) got a point compared with activity (j), then (j) has the opposite value compared to (i) Table 3. Production unit's description and performances Company Production cell Products Shifts No. 1 Yaskawa IR, two axes positioner Heat exchangers 2 No. 2 ABB IR, single axes positioner Trailer frames 2 No. 3 Yaskawa IR, single axes 2-station Forestry machine 1 (2) positioner frames Company DPP, CU, OEE, years % % No. 1 3 51 72 No. 2 2 40 70 No. 3 3 45 70 Table 4. Performance scores of main criteria Criteria CR Priority Criteria CR Priority Production unit 1.9 21 Physical properties 0 66.7 Product 24 Productivity 33.3 Company environment 55 Workforce 9.8 19.5 Cost 5.6 51.3 Performance 8.8 Maintenance 8.1 Experiences 71.7 Level 40.6 Table 5. Sub-criteria, local performance scores Criteria CR Priority Criteria CR Priority Criteria CR C1 7.4 74.3 T2 8.9 4.4 W3 2.6 C2 6.3 T3 18.1 E4 7.6 C3 19.4 T9 35.5 E9 M1 0 66.7 TE3 9.6 E10 M2 33.3 TE9 16.6 K1 L1 5.6 26 PR2 7.5 TE1 7.7 L2 41.3 E2 8.3 TE2 L3 32.7 E1 2.6 25.4 TE4 T1 5.6 41.3 E6 33.9 TE7 T4 32.7 E8 7.5 TE8 T6 26 W2 15.2 TE10 Criteria Priority C1 18 C2 23.8 C3 28 M1 8.9 M2 39.3 L1 20.2 L2 24.5 L3 20.6 T1 8.8 T4 10.1 T6 15.8 Explanations for the abbreviations are given in paragraph 5. Criteria and sub-criteria. Table 6. Suitability index results Production cell Production Product Company Suitability index unit environment Company cell No. 1 0.567 0.699 0.664 0.652 Company cell No. 2 0.494 0.709 0.804 0.716 Company cell No. 3 0.524 0.849 0.810 0.717 Table 7. Suitability decision categories Suitability index Decision Description Smaller than 0.25 No expediency Products portfolio, analysis of the current process and general conditions are indicating the lack of essential need for using robots in the company. Smaller than 0.5 To a certain extent There is indicated the strong expedient point (products, process, general conditions) and also the problematic places. The final decision lays on the industrial expert. Smaller than 0.75 Robotization is There are indicated some risks recommended which are not so much important. Higher than 0.75 Robotizing is Each group (product, process, feasible manufacturing conditions) has an index higher than 0.75, which gives a solid knowledge that robotization of the process would give significant benefits to the company.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||MANUFACTURING ENGINEERING|
|Author:||Kangru, Tavo; Riives, Juri; Mahmood, Kashif; Otto, Tauno|
|Publication:||Proceedings of the Estonian Academy of Sciences|
|Date:||Dec 1, 2019|
|Previous Article:||A simulation study to determine the parameters of medicine inventory policy.|
|Next Article:||Integration of autonomous vehicles and Industry 4.0.|