recent-ijietap.pdf

International Journal of Industrial Engineering, 19(1), 1-13, 2012.

ISSN 1943-670X INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERING

INTERNET USER RESEARCH IN PRODUCT DEVELOPMENT: RAPID AND LOW COST DATA COLLECTION

A.Shekar and J.McIntyre

School of Engineering and Advanced Technology

Massey University, Auckland New Zealand

Small to Medium Enterprises (SMEs) face enormous financial risks when developing new products. A key element of risk minimization is an early emphasis on gathering information about the end users of the product quickly. SMEs are often overwhelmed by the prospect of expected research costs, lack of expertise, and financial pressures to rush to market. Too often the more conventional path is chosen, whereby a solution is, developed and tested in the market to see if it sticks. Such methodologies are less effective and subject the SME to increased financial risk. This study demonstrates how SMEs can make use of freely available internet resources to reproduce aspects of more sophisticated customer research techniques. Internet resources such as the YouTube and Forums enable SMEs to research customers rapidly, and in a cost effective manner. This study examines New Zealand SMEs and presents two case studies to support the use of modern web-based user research in new product development. Keywords: product development, user information, web research, New Zealand

(Received 27 October 2010; Accepted in revised form 24 June 2011) 1. INTRODUCTION Small and Medium Enterprises (SMEs) are a large and vital component of most developed nations economies. The prevalence of such firms is so large that in sectors such as manufacturing, their numbers often dominate the economic landscape (Larsen and Lewis 2007). Their accrued success contributes substantially to employment, exports, and Gross Domestic Product (GDP). The sheer quantity of firms and their individual contributions build flexibility and robustness into a nations economy. Governments generally recognize this fact (Massey 2002) and support innovation in SMEs through funding research and incentive programs. The ability to launch new products and services is a critical element of success for all companies, large and small. Launching a new product or service is often the most significant financial risk a firm may face since its own inception. New product launches are typically characterized by large expenditures associated with research, production tooling, marketing and promotions. The successful recovery of expenditures and the prospect of generating profits depend entirely upon the products success in the consumer marketplace. The losses incurred from a failed product can be devastating for the small organisation. In one study of SMEs based in the Netherlands, 40% of firms were found not to survive their first 5 years in business (Vos, Keizer et al. 1998). Surveys of NZ SMEs indicate that the risks are well understood; however, NPD is still identified as a weakness within their organisation (McGregor and Gomes 1999). 1.1 SME Challenges and New Product Development Innovation poses inherent risks, yet remains an essential activity of businesses both large and small (Boag and Rinholm 1989). While SMEs are typically described as being more entrepreneurial risk-takers than their larger counterparts, in reality their situation may be more precarious. Small businesses are often more sensitive to the risks of new product development (NPD) activities due to limited financial resources. Indeed, an unsuccessful product introduction can spell disaster for the small business. While structured approaches have been successfully implemented in larger firms, smaller organisations are found to be less enthusiastic about incorporating them and struggle to adopt and make use of them (Enright 2001). The reasons for this are varied and not well understood. Many SMEs operate without the benefit of academic partnerships and may simply not be aware of the information available. Others may recognize that structured NPD approaches generally cater to the specific needs of larger firms and the results may impose unnecessary bureaucracy on the smaller organisations. It is generally recognized that smaller firms are distinct in both principle and practice from their larger counterparts. Successful large firms deal efficiently with multiple project ideas, communications involving large numbers of participants, and documentation to retain and share corporate knowledge. Smaller firms participating in the NPD process face different challenges. SMEs typically address smaller numbers of projects, involving fewer participants, and enjoy opportunities for more frequent face to face communications. Challenges to successful NPD efforts are the results of operating constraints and the culture found within smaller organisations. A partial summary of the unique issues faced by SMEs is presented in Table 1.



AN INVESTIGATION OF INTERNAL LOGISTICS OF A LEAN BUS ASSEMBLY SYSTEM VIA SIMULATION: A CASE STUDY

Aric Johnson11, Patrick Balve2, and Nagen Nagarur1

1Department of Systems Science and Industrial Engineering

Binghamton University P.O. Box 6000

Binghamton, NY. 13902-6000, USA 2Production und Logistics Department

Heilbronn University 39 Max-Planck-Strae, Heilbronn 74081, Germany

Corresponding authors email: {Aric Johnson, [email protected]}

This study involves the internal logistics of a chosen bus assembly plant that follows a lean assembly process dictated by takt time production. The assembly system works according to a rigid sequential order of assembly of different models of buses, called the String of Pearls. The logistics department is responsible for supplying kitted components to assembly workstations for the right model at the right time. A simulation model was developed to study this assembly system, with an objective of finding the minimum number of kit carts for multiple production rates and kitting methods. The implementation of JIT kitting was the ultimate goal in this case. The research focused on a specific assembly plant and therefore, the numerical results are applicable to the selected plant only. However, some of the trends in the output may be generalized to any assembly plant of similar type. Significance: This study illustrates the use of simulation to plan further lean transformation within a major bus assembly plant. This assembly plant had recently transformed their assembly operations according to lean principles with much success. The next step was to transform the logistical support to this system, and this was planned via simulation. This paper makes an original contribution to this area of research, and to the best of the authors knowledge such a work has not been published so far. Keywords: Bus assembly, kitting, takt time, simulation, internal logistics, JIT

(Received 21 March 2011; Accepted in revised form 12 March 2012) 1. INTRODUCTION Automotive industries, including bus assemblies, have been forced to cut costs to remain competitive in a global environment. For customers, price is often an important criterion, and so automotive plants strive to cut costs, while at the same time struggle to improve their throughput. The industry has mostly adopted lean manufacturing methods as the means of reducing costs and increasing throughput. Auto plants typically follow an assembly-line type of manufacturing, in which all the operations are done in stations or cells connected sequentially with a set of operations assigned for each station. This is because there are a large number of operations that need to be completed to produce a finished automobile; breaking the operations into stations allows the system to operate more efficiently and at a much faster rate. Most plants also implement a balanced assembly line of workstations that allows assemblies to flow through the system at a specific, predetermined rate, termed takt time. This balanced, sequential workstation design promotes a smooth flow throughout the plant. However, this type of system then inherits a new challenge of physically getting the required parts to the workstations on time. This problem can be described as a problem of internal logistics between parts storage (warehouse) and the many workstations. A well-coordinated logistics system is vital since a single workstation that does not receive its required parts/components on time results in delaying the entire assembly line. An assembly plant operating at a takt time production rate has little or no slack built into its schedule. Hence, getting the required parts/components to the right workstation at the right time is critical in this setting. One internal supply strategy would be to stage required parts at the workstations and replenish them as necessary. This is often not feasible in bus assembly. For one thing, the parts may be of large size and storage of such parts at a workstation may be prohibitive. In addition, if the line is producing multiple models, storing all the combinations of parts makes it more complex and tends to become more error prone. Hence, the standard practice under such situations is to let the product flow through the stations and have the parts/components for an individual assembly be brought to the appropriate workstation at the exact time they are needed. The majority of the parts/components are stored in a warehouse, and the required set of



RESEARCH-BASED ENQUIRY IN PRODUCT DEVELOPMENT EDUCATION: LESSONS FROM SUPERVISING UNDERGRADUATE

FINAL YEAR PROJECTS

A. Shekar

School of Engineering and Advanced Technology Massey University, Auckland

New Zealand

(Received 27 October 2010; Accepted in revised form 24 June 2011) This paper presents an interesting perspective on enquiry-based learning by engineering students through a project and research-based course. It discusses the lessons learned from coordinating and supervising undergraduate research-based project courses in Product Development engineering, at Massey University in New Zealand. Research is undertaken by students at the start and throughout the development project in order to understand the background and trends in the literature and incorporate them in their projects. Further research is done regarding the products technologies, problem and motivation behind the development, as well as a thorough knowledge of the context and user environment are undertaken. The multi-disciplinary nature of product development, requires students to research widely across disciplinary borders, and then to integrate the results for the goals of designing a new product and journal-style research papers. The Product Development process is a research-based decision-making process and one that needs an enquiring mind and an independent learning approach, as often the problems are open-ended and ill-defined. Both explicit and tacit knowledge are gained through this action-research methodology of learning. Tacit knowledge is gained through the hands-on project experience, experimentation, and learning by doing. Several implications for educators are highlighted, including the need for a greater emphasis on self-learning through research and hands-on practical experience, the importance of developing student research skills, and the value of learning from peer interaction. Keywords: Product development, research-based enquiry, project-based learning.

(Received 1 May 2009; Accepted in revised form 1 June 2010)

1. INTRODUCTION Engineering design programs are increasingly aware, that the project-based approach results in the development of competencies that are expected by employers (DeVere, 2010). One of these competencies is independent research skills and learning. Several new design-engineering programs have emerged and many see the need for engineers to demonstrate design and management (business) thinking in addressing product design problems. Most of these programs build the curriculum by combining courses from business, design and engineering faculties, leaving the integration to the students. We have found that this integration does not take place well. Often students tend to compartmentalise papers, do not appreciate the links between papers or sometimes lecturers from other departments are not aware of how engineers may use some of the material they cover, hence may not provide relevant examples. Hence project-based learning is an attempt to address this issue. A broad definition of project-based learning (PBL) given by Prince and Felder is: Project-based learning begins with an assignment to carry out one or more tasks that lead to the production of a final producta design, a model, a device or a computer simulation. The culmination of the project is normally a written and/or oral report summarizing the procedure used to produce the product and presenting the outcome. In practice, many engineering education activities developed on the basis of inductive instructional methods active research, inquiry-led learning and problem-based learning focus on a fixed deliverable and therefore fall within this definition of PBL. Massey University is currently reorganizing the curriculum towards overcoming the gap between theory and practice, the lack of good integration of disciplines and taking on a more student-centred approach to learning. Students follow courses in engineering sciences, physicals, mathematics, statistics and the like; however in tackling practical design projects, they fail to apply this knowledge to the extent that their design would benefit. The new curriculum proposes to have more project-based learning and less of the traditional chalk and talk teacher centred approach in all of the majors offered. This approach follows worldwide trends in engineering education, and has already been practiced within the current product development major with success, hence is presented in this paper.



A HYBRID BENDERS/GENETIC ALGORITHM FOR VEHICLE ROUTING AND SCHEDULING PROBLEM

Ming-Che Lai 1, Han-Suk Sohn 2, Tzu-Liang (Bill) Tseng 3, and Dennis L. Bricker 4

1 Department of Marketing and Logistics Management, Yu Da University, Miao-Li County 361, Taiwan

2 Dept. of Industrial Engineering, New Mexico State University, Las Cruces, NM 88003, USA 3 Dept. of Industrial Engineering, University of Texas, El Paso, TX 79968, USA

4 Dept. of Mechanical and Industrial Engineering, The University of Iowa, Iowa City, IA 52242, USA

Corresponding author: Han-Suk Sohn, [email protected]

This paper presents an optimization model and its application to a classical vehicle routing problem. The proposed model is exploited effectively by the hybrid Benders/genetic algorithm which is based on the solution framework of Benders decomposition algorithm, together with the use of genetic algorithm to effectively reduce the computational difficulty. The applicability of the hybrid algorithm is demonstrated in the case study of the Rockwell Collins fleet management plan. The results demonstrate that the model is a practical and flexible tool in solving realistic fleet management planning problems. Keywords: Vehicle Routing, Hybrid Algorithm, Genetic Algorithm, Benders Decomposition, Lagrangian Relaxation, Mixed-integer programming.

(Received 9 June 2011; Accepted in revised form 28 February 2012) 1. INTRODUCTION The vehicle routing problem (VRP) involves a number of delivery customers to be serviced by a set of identical vehicles at a single home depot. The objective of the problem is to find a set of delivery routes such that all customers are served exactly once and the total distance traveled or time consumed by all vehicles is minimized, while at the same time the sum of the demanded quantities in any routes does not exceed the capacity of the vehicle. The VRP is one of the most challenging combinatorial optimization problems and it was first introduced by Dantzig and Ramser (1959). Since then, the VRP has stimulated a large amount of researches in the operations research and management science community (Miller, 1995). There are substantial numbers of heuristic solution algorithms proposed in the literature. Early heuristics for this problem are those of Clarke and Wright (1964), Gillett and Miller (1974), Christofides et al. (1979), Nelson et al. (1985), and Thompson and Psaraftis (1993). A number of more sophisticated heuristics have been developed by Osman (1993), Thangiah (1993), Gendreau et al. (1994), Schmitt (1994), Rochat and Taillard (1995), Xu and Kelly (1996), Potvin et al. (1996), Rego and Roucairol (1996), Golden et al. (1998), Kawamura et al. (1998), Bullnheimer et al. (1998 and 1999), Barbarosoglu and Ozgur (1999), and Tom and Vigo (2003). As well, exact solution methods have been studied by many authors. These include branch-and-bound procedures, typically with the basic combinatorial relaxations (Laporte et al., 1986; Laporte and Nobert, 1987; Desrosiers et al., 1995; Hadjiconstantinou et al., 1995) or Lagrangiran relaxation (Fisher, 1994; Miller, 1995; Toth and Vigo, 1997), branch-and-price procedure (Desrochers et al., 1992), and branch-and-cut procedure (Augerat et al, 1995; Ralphs, 1995; Kopman, 1999; Blasum and Hochstattler, 2000). Unlike many other mixed-integer linear programming applications, however, Benders decomposition algorithm was not successful in this problem domain because of the difficulty of solving the master system. In mixed-integer linear programming problems, where Benders algorithm is most often applied, the master problem selects values for the integer variables (the more difficult decisions) and the subproblem is a linear programming problem which selects values for the continuous variables (the easier decisions). For the VRP problem, the master problem of Benders decomposition is more amenable to solution by a genetic algorithm (GA) which searches the solution space in parallel fashion. The fitness function of the GA is, in this case, evaluated quickly and simply by evaluating a set of linear functions. In this short paper, therefore, a hybrid algorithm is presented in order to overcome the difficulty in implementing the Benders decomposition for the VRP problem. It is based on the solution framework of Benders decomposition algorithm, together with the use of GA to effectively reduce the computational difficulty. The rest of this paper is organized as follows. In section 2 the classical vehicle routing problem is presented. The application of the hybrid algorithm is described in section 3. In Section 4, a case study on the fleet management planning of the Rockwell Collin, Inc. is presented. Some concluding remarks are presented in Section 5. Finally, Section 6 lists references used in this paper.



A NEW TREND BASED APPROACH FOR FORECASTING OF ELECTRICITY DEMAND IN KOREA

Byoung Chul Lee, Jinsoo Park, Yun Bae Kim

Department of Systems Management Engineering, Sungkyunkwan University, Suwon, Republic of Korea

Corresponding author: Yun Bae Kim, [email protected]

Many forecasting methods for electric power demand have been developed. In Korea, however, these kinds of methods do not work correctly. A peculiar seasonality in Korea increases the forecasting error produced by previous methods. Two big festivals, Chuseok and Seol, also produce forecasting errors. Therefore, a new demand forecasting model is required. In this paper, we introduce a new model for electric power demand forecasting which is appropriate to Korea. We start the research using the concept of weekday average. The final goal is to forecast hourly demand for both the long and short term. We finally obtain the result with accuracy of over 95%. Keywords: Demand forecasting, electric power, moving average

(Received 7 April 2010; Accepted in revised form 24 June 2011) 1. INTRODUCTION There have been many studies related to forecasting electric demand. These studies have contributed to achieving greater accuracy. Shahiderhpour et al. (2002) introduced market operation in electric power systems. Price modeling for electricity markets was described by Bunn (2004). Kawauchi et al. (2004) developed a forecasting method based on conventional chaos theory for short term forecasting. Gonzalez-Romera et al. (2007) used neural network theory, Oda et al. (2005) forecasted demand with regression analysis, and Pezzulli et al. (2006) focused on seasonal forecasting with a Bayesian hierarchical model . These attempts, while valuable, are inappropriate for Korea because there are four distinct seasons, which have their own feature such as a cycle of three cold days and four warm days in winter. In addition, Korean demand trend has a cycle by weekly unit. Therefore we have two sources of seasonality, seasonal factor and weekly factor. Therefore, the previous methods are not proper due to double seasonality. To examine double seasonality, we analyzed past data to determine properties of Korean electric demand. Using these properties, we defined a new concept of weekday average, (WA), and developed models for forecasting hourly demand of electric power in Korea. The organization of this paper is as follows. In Section 2, the concept of WA is used for 24 hours as the first step in forecasting hourly demand. In Section 3, we deal with the methods of forecasting WA and non-weekday demand, including holidays and festivals. We apply our model to the actual demand data and show the results in Section 4. We conclude the research and suggest further studies in Section 5.

2. CONCEPT OF WEEKDAY AVERAGE We found two special properties related to the hourly demand of electric power in Korea; one is the character of weekdays, and the other is a connection between weekdays and non-weekdays (a weekday means the days from Tuesday to Friday). Holidays and festival seasons are regarded as non-weekdays even though they are in a weekday period. The demands during each weekday are almost similar to one another at the same hour; this is the first property. However, the demands of Monday and weekends are less than those of weekdays by an invariable ratio; this is the second property. Therefore, our research starts by developing a method for forecasting the hourly demand of weekdays. We then find the relation between weekdays and non-week days. Let us define the hourly demand:

)(hD in : demand from 00:)1( h to 00:h

24,,1( =h and ).7,,1=i

... (1)

where n is the number of weeks from the base week; for example, if the base week is the first week in 2007, then Dec. 31st, 2007 has the value of 52=n . i is the day of the week (1=Monday, 7= Sunday).



RELIABILITY EVALUATION OF A MULTISTATE NETWORK UNDER ROUTING POLICY

Yi-Kuei Lin

Department of Industrial Management

National Taiwan University of Science and Technology Taipei, Taiwan 106, R.O.C.

Tel: +886-2-27303277, Fax: +886-2-27376344 Corresponding author: Lin, [email protected]

A multistate network is a stochastic network composed with multistate arcs in which each arc has several possible capacities and may fail due to failure, maintenance, etc. Different from the deterministic case, the minimum transmission time in a multistate network is not a fixed number. We evaluate the probability that a given amount of data/commodity can be sent from a source port to a sink port through a pair of minimal path (MP) simultaneously under the time constraint. Such a probability is named the system reliability. An efficient solution procedure is first proposed to calculate it. In order to enhance the system reliability, the network administrator decides the routing policy in advance to indicate the first and the second priority pairs of MP. Subsequently, we can evaluate the system reliability under the routing policy. An easy criterion is then proposed to derive an ideal routing policy with higher system reliability. We can treat the system reliability as a performance index to measure the transmission ability of a multistate network such as computer, logistics, urban traffic, telecommunication systems, etc. Keywords: Multistate network; commodity transmission; system reliability; transmission time; routing policy

(Received 1 March 2010; Accepted in revised form 27 February 2012) 1. INTRODUCTION For a deterministic network in which each arc has a fixed length attribute, the shortest path problem is to find a path with minimum total length. When commodities are transmitted from a source to a sink through a flow network, it is desirable to adopt the shortest path, least cost path, largest capacity path, shortest delay path, or some combination of multiple criteria (Ahuja, 1998; Bodin et al., 1982; Fredman and Tarjan, 1987; Golden and Magnanti, 1977), which are all variants of the shortest path problem. From the point of view of quality management and decision making, it is an important task to reduce the transmission time through a flow network. Hence, a version of the shortest path problem called the quickest path problem proposed by Chen and Chin (1990) arises. This problem finds a quickest path with minimum transmission time to send a given amount of data/commodity through the network. In this problem, each arc has the capacity and the lead time contributes (Chen and Chin, 1990; Hung and Chen, 1992; Martins and Santos, 1997; Park et al., 2004). More specifically, the capacity and the lead time are both assumed to be deterministic. Several variants of quickest path problems are thereafter proposed; constrained quickest path problem (Chen and Hung, 1994; Chen and Tang, 1998), the first k quickest paths problem (Chen, 1993; Chen, 1994; Clmaco et al., 2007; Pascoal et al., 2005), and all-pairs quickest path problem (Chen and Hung, 1993; Lee and Papadopoulou, 1993). However, due to failure, partial failure, maintenance, etc., each arc should be considered as multistate in many real-life flow networks such as computer, logistics, urban traffic, telecommunication systems, etc. That is, each arc has multiple possible capacities or states (Jane et al., 1993; Lin et al., 1995; Lin, 2003, 2004, 2007a,b, 2009; Yeh, 2007, 2008). Then the transmission time thorough a network is not a fixed number if each arc has the time attribute. Such a network is named a multistate network throughout this paper. For instance, a logistics system with each node representing the shipping port and each arc representing the shipping itinerary between two ports is a typical multistate network. The capacity of each arc is counted in terms of number of container, and is stochastic due to that either containers or traffic tools (e.g., cargo airplane, cargo ship, etc.) through each arc may be in maintenance, reserved by other suppliers or in other conditions. The purpose of this paper is to design a performance index to measure the transmission ability for a multistate network. In order to reduce the transmission time, the data/commodity can be transmitted through several minimal paths (MPs) simultaneously, where an MP is a sequence of arcs without loops. For convenience, we first concentrate on commodity transmission through two MPs. We mainly evaluate the probability that the multistate network can send d units of commodity from a source port to a sink port through a pair of MP under the time constraint T. Such a probability is named the system reliability, which can be treated as a performance index. Under the same time constraint and demand requirement, the system owns a better transmission ability if it obtains the higher system reliability. In order to boost the transmission ability, the network administrator decides the routing policy in advance to indicate the first and the second

International Journal of Industrial Engineering, 19(2), 68-79 2012.


DETERMINING THE CONSTANTS OF RANGE CHART FOR SKEWED POPULATIONS

Shih-Chou Kao

Graduate School of Operation and Management, Kao Yuan University, No.1821, Jhongshan Rd., Lujhu Dist., Kaohsiung City 821, Taiwan (R.O.C.).

Corresponding author email: [email protected] The probability of a false alarm rate (type I risk) in Shewhart control charts based on a normal distribution will increase as the skewness of a process increases. However, the distribution of a range is a positivelyskewed one. It is unstable for monitoring range values by using three-sigma control limits that is from the concept of a normal assumption. Moreover, most studies employ a simulation method to compute the type I risks of the range control chart for nonnormal processes. To provide an alternative method, this study utilizes the probability density function of the distribution of the range to construct the appropriated control limits of a range control chart for a skewed process. The control limits of the range chart were determined by setting that the type I risk is equal to 0.0027 and the standardized Weibull, lognormal and Burr distributions. Furthermore, compared to range charts that use type I risks and type II risks, weighted variance (WV), skewness correction (SC) and traditional Shewhart control charts, the proposed range chart is superior to other control chart, in terms of the type I risks and type II risks for a skewed process. An example of the yield strength for the deformed bar in coil is presented to illustrate these findings. The study utilized the probability density function of range distribution and =0.0027 probability limits with considering the three distributions, Weibull, lognormal and Burr to construct the R control chart. The computed constants of the R control chart were listed in a table that can be consulted by for practitioners. R chart using the proposed method is superior to other control chart, in terms of the type I risks and type II risks for a skewed process. Keywords: Range chart, skewed distribution, normality, type I risk.

(Received 22 March 2010; Accepted in revised form 24 June 2011) 1. INTRODUCTION The development of control charts became rapid and diverse after W. A. Shewhart proposed a traditional control chart. Control charts have the superior ability for monitoring a process in manufacturing, and they have been applied successfully in other areas, such as finance, health care and information. The Shewhart range (R) control chart is one of the most frequently used control charts since it is easily operated and interpreted by practitioners. In general, traditional variable control charts, such as an average and a R control charts, are based on the normality assumption. However, many processes in industry violate this assumption. These skewed processes involve chemical processes, cutting tool wear processes and lifetime in an accelerated life test (Bai and Choi, 1995). Moreover, the range distribution is a positivelyskewed one (Montgomery, 2005). If the traditional control charts are used to monitor a nonnormal process, the probabilities of a type I error () in the control charts increases as the skewness of the process increases (Bai and Choi, 1995; Chang and Bai, 2001). Bai and Choi (1995), Chang and Bai (2001) and Montgomery (2005) considered four methods for improving the capabilities of control charts for monitoring a skewed process. The first method increased the sample sizes on the basis of the central limit theorem. When the samples are larger, the skewed distribution will become a normal or approximately normal distribution. However, the method is often expensive due to sampling. The second method is to assume that the distribution of a process is known and then to derive a suitable control chart from this known distribution. Ferrell (1958) designed geometric midrange and range control charts for a lognormally distributed process. Nelson (1979) proposed median, range, scale and location control charts for a Weibull distribution. The third method is to construct the traditional control chart using approximately normal data that result from transforming skewed data. Various criteria were proposed to transform exponential data, such as maximum likelihood and Bayesian methods (Box and Cox, 1964), KullbackLeibler (KL) information numbers (Hernandez and Johnson, 1980; Yang and Xie, 2000), measure of symmetry (zero skewness; Nelson, 1994), ease of use (Kittlitz, 1999) and minimizing the sum of the absolute differences (Kao et al., 2006), to assess transformation efficiency. The shortcoming of this method is that it is difficult to identify an exact distribution of a process with the second method. The last method is to construct control charts using heuristic methods with no assumption on the form of the distribution. Choobineh and Ballard (1987) proposed the WV method to determine the constants of average and R charts based on the semivariance estimation of Choobineh and Branting (1986). Bai and Choi (1995) considered the three skewed distributions (Weibull, lognormal and Burr) and determined the constants of average and R charts using the weighted variance (WV) method by splitting a skewed distribution in two parts at the mean. Chang and Bai (2001) decided the constants of average control chart by replacing a variance of WV method with a standard deviation. Chan and Cui (2003) proposed the skewness correction (SC) method based on the CornishFisher expansion (Johnson et al.,

International Journal of Industrial Engineering, 19(2), 80-89 , 2012.


PERFORMANCE MODELING AND AVAILABILITY ANALYSIS OF SOLE LASTING UNIT IN SHOE MAKING INDUSTRY: A CASE STUDY

Vikas Modgil 1, S.K. Sharma2, JagtarSingh3

1Dept of Mechanical Engineering, D.C.R.U.S.T., Murthal, Sonepat, Haryana, India

2Dept of Mechanical Engineering, N.I.T Kurukshetra, Haryana, India 3Dept of Mechanical Engineering, S.L.I.E.T Longowal, Sangrur, Punjab, India

Corresponding author: Vikas Modgil, [email protected]

In the present work Performance modelling of the sole lasting unit, a part of shoe making industry has been done on the basis of Markov birth-death process using probabilistic approach for the purpose to compute and improve the time dependent system availability (TDSA). The kolmogorov-differential equations based on mnemonic rule are formulated using the performance model and are solved to estimate the availability of the system as a function of time month wise for the whole year using a more sensitive and advance numerical technique, known as adaptive step-size control Runge-Kutta method. The input contributors for the computation of time dependent system availability of the system are the existing failure and repair rate are taken from plant maintenance history sheets. The new repair rates are also devised for the purpose of maximum improvement in the availability. The analysis finding helps the plant management for adapting the best possible maintenance strategies. Performance modeling and availability analysis of a practical system is conducted in the paper with the purpose to improve its operational availability. The time dependent system availability (TDSA) is computed with the existing failure and repair rates on the monthly basis for the whole year. New devised repair rates are also proposed through which one can assure maximum availability of the system with existing equipments/or machines. It is also explored that the, the knowledge of TDSA minimizes the chances of sudden failure and assure the maximum availability of the system and exposes the critical subsystems which needs more attention and due consideration as far as the maintenance is concerned. The improvement in the availability of the system is mostly from 2% to 5% in most of the month. However it increases drastically to 9% in the month of April. Further the assured increase in availability increases productivity as well as the balance between demand and supply such that the manufacturer delivers its product properly in time to the market/society, which in turn increases the profit and the reputation of industry in the market.

Keywords: Performance Modelling, Time Dependent System Availability (TDSA); Runge-Kutta; Sole Lasting. Kolmogorov-Differential equation, Shoe Making.

(Received 25 September 2011; Accepted in revised form 27 February 2012)

1. INTRODUCTION With increasing advancement and automation, the industrial systems are getting complex and thus maintaining their failure-free operation is not only costly but also difficult. Thus maximum availability levels are desirable to reduce the cost of production and maintaining them in working order for a long duration. The industrial operating conditions and repair facility play also an important role in this regard. Several attempts have been made by various researchers and authors to find the availability of practical industrial system using different techniques. Dhillon and Natesan (1983) examined the availability of power system in fluctuating environment. Singh I.P. (1989) studied the reliability analysis of a complex system having four types of components with pre-emptive priority repairs. Singh and Dayal (1992) studied the reliability analysis of a repairable system in a fluctuating environment. Gupta et al. (2005) evaluated the reliability parameters for butter manufacturing system in a diary plant considering exponentially distributed failure rates of various components. Solanki et al (2006) evaluated the reliability of thermal-hydraulic passive systems using thermal hydraulic code RELAP 5/MOD 3.2(which operate in two phase natural circulation). Rajpal et al (2006) employed artificial neural network for modelling reliability, availability and maintainability of a repairable helicopter transport facility. Kumar et al. (2007) developed a simulated availability model for CO2 cooling system of a fertilizer plant. Goyal et al. (2009) discusses the steady state availability analysis of a part of rubber tube production system under pre-emptive priority repair using Laplace transform technique. Garg S.et al. (2010) computed the availability of crank-case manufacturing in a 2-wheeler automobile industry and block board system under pre-emptive priority discipline. In this paper a sub-system of the practical plant Liberty Shoes Limited which is a continuous production system is taken and the time dependent system availability of the system is estimated using a more advance and sensitive numerical technique known as adaptive step-size runge-kutta method. The earlier work carried out by most of the research groups do not entertain this aspect of time dependent availability. They just provide the long run or steady state availability of the system by taking time infinity. The liberty shoe making plant situated in Karnal, Haryana, India is chosen for study. Numerical results based upon the true data collected from industry are presented to illustrate the



SIMULATION MODELING OF OUTBOUND LOGISTICS OF SUPPLY CHAIN: A CASE STUDY OF TELECOM COMPANY

Arvind Jayant1, S. Wadhwa2, P.Gupta3, S.K.Garg4

1,3Department of Mechanical Engineering

Sant Longowal Institute of Engg. & Technology, Longowal, Sangrur, Punjab 148106 (INDIA) 2Department of Mechanical Engineering, Indian Institute of Technology, Delhi (INDIA) 4Department of Mechanical Engineering, Delhi Technological University, Delhi-110042

Corresponding author: Arvind Jayant, [email protected]

The present work has been done for a telecom company with a focus on cost and flexibility in effectively deals with changing scenario. In this paper, the major problems faced by company at upper end of supply chain and sales outlet are analyzed and a complete inventory analysis on one of a company product is done by developing an Inventory model for the company bound store/distribution center and optimal inventory policy is suggested for the outbound logistics on the basis of simulation analysis. This model is flexible enough to respond to the market fluctuations more efficiently and effectively. The model is developed in Microsoft EXCEL. Significance: Increasing competitive pressures and market globalization are forcing the firms to develop supply chains that

can quickly respond to customer needs. The inventory model for the companys bound store/outbound logistics has been developed & simulated to reduce the operating cost, stock out, to make supply chain agile.

Key words: Supply Chain, Outbound Logistics, Information Technology, Simulation, Operating Cost, Inventory.

(Received 4 August 2010; Accepted in revised form 28 February 2012) 1. INTRODUCTION

The basis of global competition has changed. No longer are companies competing against other companies, but rather supply chains are competing against supply chains. Indeed, the success of a business is now invariably measured neither by the sophistication of its product nor by the size of the market share. It is usually seen in the light of the ability to sometimes forcefully and deliberately harness its supply chain to deliver responsively to the customers as and when they demand it. Flexible Supplier-manufacturer relationship is the key enabler in the supply chain management, without the flexibility at the vendor side the supply chain cant respond fast. Therefore, the relationship with the supplier should be flexible enough to meets the changing market needs [2]. In this paper several experiments were carried out on the model for visualizing the impact of the various decision variables on the total cost and then fixing up the values of (s) and (S). The graphs showing the impact of these parameters on the performance of the individuals and the system were plotted. Based on the systems performance under different sets of operating decisions we shall try to analyze the effect of the different parameters and in what manner their decisions affect the performance of others across the chain. The parameters whose impact was studied are stock level (S), reorder level (s); this paper deals with the impact of increase in stock levels and reorder level of the warehouse on overall system performance [6]. 2. ABOUT THE PRODUCT Bharti -Teletech is a giant in the manufacturing of all kind of telephone sets for the Department of Telecommunication, open market and for exports. The company share in this segment is highest in India. This company has 35% share in telephone segment in India. The company is producing the seven model of telephone with brand name of beetal.

The company is currently facing the problem of delivering the CORAL & MILL I model of phones on schedule date. Though the number of shortage is small but any delivery made beyond schedule will be considered as the lost opportunity of sale.

The coral is general model for the open market and its demand is highly uncertain therefore frequent stock outs are going on at the end of bound store and warehouse side.

The forecasts generated using the 6-month average were not giving the appropriate results. The warehouse is not using the any inventory policy and the reorder level of the warehouse was made intuitively

made.



SIMULATION-BASED OPTIMIZATION FOR RESOURCE ALLOCATION AT

THIRD-PARTY LOGISTICS SYSTEMS Yanchun Pan1, Ming Zhou2, Zhimin Chen3

1,3College of Management, Shenzhen University, P.R. China 2Center for Systems Modeling and Simulation, Indiana State University

Corresponding author: Ming Zhou, [email protected] Allocating resource at third-party logistics systems differs significantly from traditional private logistics systems. The resources are considered commodities sold to customers of different types. Total yield suffers when over-allocate to lower-rate or price-sensitive customers; but the resource become spoiled when reserve too much for full-rate or time-sensitive customers that do not arrive as expected. Uncertain order characteristics make the optimization of such decisions very hard, if not impossible. In this paper we proposed a simulation-based optimization to address related issues. A genetic algorithm based optimization module is developed to generate/search good solutions; and a discrete-event simulation model is created to evaluate the solutions generated. The two modules are integrated to work in evolutionary cycles to achieve the optimization. The study also compared GA/Simulation model with more traditional approach such as response surface methodology via designed experiments. The models were validated through experimental analysis. Keywords: resource allocation; simulation; genetic algorithm; optimization; third-party logistics

(Received 2 September 2010; Accepted in revised form 1 March 2012)

1. INTRODUCTION

Studies on third-party logistics (TPL) systems have been thriving since last two decades, as TPL systems gain popularity in many parts of the world through the flexibility and convenience they provide to improve the quality and efficiency of logistics services and customer satisfaction (Lambert et al, 1998; Bowersox et al, 2002). Resource or capacity allocation (e.g. allocation of warehouse space for temporary storage of customer goods) at TPL systems differs significantly from traditional private logistics system. Unlike private systems, TPL companies use public warehouses that are usually more efficient than private ones through better productivity, shared resources, economy of scale, and transportation (delivery) consolidation (Ackerman, 1994); and consider the resources to be allocated as commodities sold directly to different customers repeatedly via services generated based on the resources, such as storing, handling, or transporting goods. Also such resources are considered perishable when they are not sold at or during a period of time, i.e. they cause the loss of possible revenue that could have been otherwise generated if they were sold (Phillips, 2005). As in airline or hospitality industries, there are mainly two types of customer demands, and accordingly two different approaches for allocating resource to customer orders. First, many customers prefer to have their orders placed in advance a period of time to expect a discounted rate of service. Once allocated, the chunk of resource is locked in and subtracted (from available stock) for the usage period of the order, which is a time period during which the allocated resource is consumed to generate service for the order. This type of customer is price-sensitive. The risk of over-allocating resource to this kind of orders is that we may lose opportunities to serve more profitable full-rate customers (or customers willing to pay higher rates). This is known as the risk of spill (Humphreys, 1994; Phillips, 2005). On the other hand, there are customers who are less price-sensitive, but more time-sensitive, i.e. they place orders often at a time very close to (or at) the



TRACKING AND TRACING OF LOGISTICS NETWORKS: PERSPECTIVE OF REAL-TIME BUSINESS ENVIRONMENT

AHM Shamsuzzoha and Petri T Helo

Department of Production University of Vaasa, PO BOX 700, FI-65101, Finland

Todays business environments are full of complexities in terms of managing the value adding supply chain and logistics networks. In recent years, the development of locating and identifying technologies contribute to fulfill the growing demands of tracking and tracing the logistics and/or transportation chain. The importance of tracking and tracing of shipments is considered quite high for manufacturing firms in respect to managing logistics networks efficiently and satisfying high customers demand. This paper presents a theoretical overview of sophisticated technology-based methodology or approach required for solving the complex tracking and tracing system in the logistics and supply chain network. A real-life case example is presented in this paper with the view to demonstrate the tracking technology in terms of identifying the location and related conditions of the case shipment. The overall outcomes from this research are concluded with future research direction too. Significance: This work basically reviews the existing tracking and tracing technologies available over the areas of logistics and supply chain management. It also demonstrates the methodology for implementing such technologies in real-life business cases and provides insight of tracking and tracing technology with respect to identifying location, position and conditions of the shipped items. Keywords: Logistics tracking and tracing, IT-based solution, Transportation and Distribution network, Real-time information flow, Business competition.

(Received 3 June 2011; Accepted in revised form 31 July 2011) 1. INTRODUCTION The identification of location and knowing the conditions of the transported items on real-time business environment are growing increasing concern in todays business. This is very much expected for the manufacturing firms in terms of their business growth and making the customers happy. The importance of tracking and tracing of shipments is considered quite high for manufacturing firms in terms of customer service and essential for managing logistics networks efficiently. Global industries are facing problems both from tracking and tracing in their logistics networks that creates huge coordination problems in the overall product development sites. This problem looses the track among production, delivery and distribution in the complete logistics chain from source to destination, which is responsible for opportunity cost through customers dissatisfaction. Tracking system helps to identify the position of the shipment and informed the customer in well advance. Without tracking system it is almost impossible to find out delivered items and often considered as lost or stolen item that causes business loss. This system might fulfill the needs of project manager to map the production process from transportation to material management (Helo et al., 2005, Helo, 2006). Recently evolved technologies supports the fundamental needs for tracking and tracing the logistics network. The tracking technology ensures the real-time status update of the target shipment and provides the detailed information corresponding to location, conditions of the shipments (vibration, damage, missing, etc). In practice, there are several tracking systems available through GPS, GTIN (EAN Int., 2001), RFID (ISO/IEC, 2000; Chang, 2011), Barcode etc; however, all these systems are not fully compatible for industry. Most of the available tracking and tracing systems utilize proprietary tracking numbers defined by the individual companies operating systems and are based on information architecture, where the tracking information is centralized to the provider of the tracking service. Existing tracking systems can not able to identify the contents within a box for example, whether the box is open or the contents are lost or stolen etc. In order to tackle such misalignments in the logistics channel, a state-of-the art technologies or tools are needed to be developed for sustainable production process. These tools are needed to be cost effective and at the same time possibility for reuse or recycling for any circumstances. Before proceed towards the real-time tracking technology, it is crucial to analyze its possible cause and effects. Optimal performance measures for the technologies could ensure projects success for any industries. Tracking technologies in logistics networks are implemented fairly little in the global technology industry. Mostly high volume of global industries are implemented this technology with limited capabilities. The basic methods for all these tracking systems are usually confined for the customer to access the tracking information are within the area of tracing the shipments through manual queries such as using a www-site or telephone call, e-mailing, fax or to engage in developing



A MATHEMATICAL PROGRAMMING FOR AN EMPLOYEES CREATIVITY MATRIX CUBIC SPACE CLUSTERING IN ORGANIZATIONS

Hamed Fazlollahtabar*1, Iraj Mahdavi2, Saber Shiripour2, Mohammad Hassan Yahyanejad3

1Faculty of Industrial Engineering, Iran University of Science and Technology, Tehran, Iran 2Department of Industrial Engineering, Mazandaran University of Science and Technology, Babol, Iran

3Mazandaran Gas Company, Sari, Iran *Corresponding authors email: [email protected]

We investigate different structural aspects teams network organization and their creativity within a knowledge development program (KDP). Initially, a pilot group of employees in an organization is selected. This group is evaluated through creativity parameters using a questionnaire. Considering the questionnaires data, a creativity matrix is configured by a binary scoring. Applying the creativity matrix, clustering is performed via mathematical programming. The pilot group is divided into some research teams. The research subjects are submitted to the teams. Finally, an allocated problem is solved and some new research subjects are evolved to be assigned to the next configured teams. This procedure is repeated dynamically for different time periods. Keywords: Creativity matrix; Intelligent clustering; Cubic space clustering

(Received 28 September 2011; Accepted in revised form 20 December 2011) 1. INTRODUCTION In todays knowledge-intensive environment, Knowledge Development Programs (KDPs) are increasingly employed for executing innovative efforts (Oxley and Sampson, 2004; Smith and Blanck, 2002). Researchers and practitioners mainly agree that effective management plays a critical role in the success of such KDPs (Pinto and Prescott, 1988). Unfortunately, the knowledge and experience base of most managers refer to smaller-scale projects consisting of only a few project teams. This may be responsible for what Flyvbjerg et al. (2003) call a performance paradox: At the same time as many more and much larger infrastructure projects are being proposed and built around the world, it is becoming clear that many such projects have strikingly poor performance records .... KDPs employ follow a project-management like approach with the team as the organizational nucleus (e.g., van Engelen et al., 2001). The information network of these teams defines the opportunities available to them to create new knowledge (e.g., Uzzi, 1996). As many scholars have argued, networks of organizational linkages are critical to a host of organizational processes and outcomes (e.g., Baum and Ingram, 1998; Darr et al., 1995; Hansen, 1999; Reagans and McEvily, 2003; Szulanski, 1996). New knowledge is the result of creative achievements. Creativity, therefore, molds the foundation for poor or high degree of performance. The extent to which teams in KDPs produce creative ideas depends not only on their internal processes and achievements, but also on the work environment in which they operate (e.g., Amabile et al., 2004; Perry-Smith and Shalley, 2003; Reiter-Palmon and Illies, 2004). Since new knowledge is mainly created when existing bases of information are disseminated through interaction between interacting teams with varying areas of expertise, creativity is couched in interaction networks (e.g., Leenders et al., 2003; Hansen, 1999; Ingram and Robert, 2000; Reagans and Zuckerman, 2001; Tsai, 2001; Uzzi, 1996). Any organization needs team work among employees for productivity purposes in problem solving. Organizations face various problems in their determined missions. A useful approach to address these problems is to configure teams consisting of expert employees. Due to their knowledge and experience of the organization, these teams understand the organization's problems better more than external research groups and thus may solve the problems more effectively. Hence, the significant decision to be made is configuration of the teams. Creative teams would be able to propose more practical and beneficial solutions for organization's problems. Since creativity is a qualitative concept, analyzing and decision making require knowledge management algorithms and methodologies. These methodologies are employed in the different steps of configuring teams, task assignment to teams, teams' progress assessment and executive solution proposals for problems. In the present work, we propose a creativity matrix analyzing creativity parameters of a pilot group in an organization. Then, using an intelligent clustering technique, research teams are configured and research subjects are allocated to them.



ACCEPTANCE OF E-REVERSE AUCTION USE: A TEST OF COMPETING MODELS

Fethi Calisir and Cigdem Altin Gumussoy

Department of Industrial Engineering Istanbul Technical University

This study aims to understand factors affecting e-reverse auction usage in companies by comparing three models: Technology Acceptance Model (TAM), Theory of Planned Behavior (TPB) and the integrated model (integration of TAM and TPB). The comparison of the models will answer two important questions: First, with the integration of the models, whether the explanation rate of behavioral intention to use and actual use is increased. Second, in explaining e-reverse auction usage, whether TAM is the most powerful method. Since TAM is developed only to explain usages of information technologies (IT). Using LISREL 8.54, data collected from 156 employees working in the procurement department of companies in 40 different countries were used to test the models. Results indicated that, TPB may be more appropriate than the TAM and the integrated model for explaining behavioral intention to use e-reverse auction. Further, the explanation rate of both behavioral intention to use and actual use is not increased with the integration of the models. The other result suggests that behavioral intention to use is explained- by only attitude towards use in TAM; by subjective norms, perceived behavioral control and attitude towards use in both TPB and the integrated model. Actual use of e-reverse auction is directly predicted by behavioral intention to use in all three models. This study concludes with the discussion of the findings, implications for practitioners and recommendations for possible future research. Significance: This paper aims to identify significant factors affecting e-reverse auction usage among buyers working in

the procurement department of companies by comparing three models: TAM, TPB and the integrated model. The comparisons will explore that whether the explanation rates of behavioral intention to use and actual use is increased with the integration of the models and whether TAM is the most powerful method in explaining the usage behavior of e-reverse auction users.

Keywords: E-reverse auction, TAM, TPB, Integrated model, Actual use, Model comparison

(Received 7 June 2011; Accepted in revised form 18 September 2011) 1. INTRODUCTION E-reverse auction is an online- and real-time auction between a buying company and two or more suppliers (Carter et al., 2004). Use of the e-reverse auction tool was first offered by FreeMarkets in 1999 and has since then been progressively adopted more intensively by firms. Several Fortune Global 2000 companies use e-reverse auction as a purchasing tool (Giampietro and Emiliani, 2007). For example, General Electric spends 50-60 billion $ per year and people in positions of responsibility believe that 50-66% of this amount can be auctioned (Hannon, 2001). Using e-reverse auction offers many advantages to buyers as well as suppliers. Price reduction is undoubtedly the most important one. Suppliers may have to make higher price reductions to win the auction (Giunipero and Eltantawy, 2004). In addition to the price advantage, increase in buyer productivity, reduction in cycle time, access to many suppliers at the same time, creating a more competitive environment, standardization, and transparency in purchasing process are the other advantages of e-reverse auction. All these advantages create more opportunities for companies by reduction in cost and time, enabling these companies can offer higher quality products (Carter et al., 2004; Bartezzaghi and Ronchi, 2003). In 2000, General Electric saved $480 million by using e-reverse auction from its $6.4 billion expenditure (Hannon, 2001). E-reverse auction has benefits not only for buyers but also for suppliers. These are growing markets, accessed by system users all over the world, who are enabled to compare their own competitiveness in the market and follow up auctions by potential customers on the Internet. Besides, they can estimate their customers needs and market trends by checking the e-reverse auctions specifications and conditions for the products and services. Thus, suppliers can not only see areas for improvement and but also their own needs for improvement (Emiliani, 2000; Mullane et al., 2001). Therefore, it is important to explain and understand the factors that affect the use of e-reverse auctions as they aim at improving performances of company and employees to complement each other. To our knowledge, the only study that compares models in the context of e-auction is Bosnjak et al. (2006). In their study, they aim to explain English auction use, which is generally used in business-to-customer and customer-to-customer markets, whereas the current study is related with e-reverse auction technology, used for procurement of products or services in the business-to-business markets.



A METHODOLOGY FOR PERFORMANCE MEASUREMENT IN MANUFACTURING COLLABORATION

Jae-Yoon Jung1, JinSung Lee1, Ji-Hwan Jung2, Sang-Kuk Kim1, and Dongmin Shin3

1Department of Industrial and Management Systems Engineering, Kyung Hee University, Korea 2Business Innovation Center, LG Display, Korea

3Department of Industrial and Management Engineering, Hanyang University, Korea

Corresponding author: Dongmin Shin, [email protected] Effective performance measures must be developed in order to effectively maintain successful collaboration. This paper presents a methodology of collaborative performance measures to evaluate the overall performance of a collaboration process between multiple manufacturing partners. The partners first define collaborative key performance indicators (cKPI), and they then measure the cKPIs and calculate the synthetic performance from the cKPI values to evaluate the result of the collaboration case. To measure different scales of cKPI, we develop a two-folded desirability function based on the logistic sigmoid functions. The proposed methodology provides a quantitative way to measure collaborative performance in order to effectively manage collaboration among partners, continuously improving collaboration performance. Keywords: Manufacturing collaboration, performance measurement, collaborative key performance indicators, two-folded desirability function, sigmoid function.

(Received 17 May 2011; Accepted in revised form 18 September 2011) 1. INTRODUCTION One important change in the manufacturing industry is that competition between individual companies has been extended to competition between the manufacturing networks surrounding the companies (NISA, 2001). This is because the competitive advantages of modern manufacturing companies are derived from manufacturing collaboration in virtual enterprise networks such as supply chains (Mun et al., 2009). Most existing performance measures, however, have been developed to evaluate the performance of internal or outsourcing projects from the perspective of a single company (Ghalayini et al., 1997; Khadem et al., 2008; Koc, 2011). Moreover, some performance indicators such as trading costs are oriented to a single company, and cannot be directly applied to measuring the collaboration performance since such indicators conflict between two partners. As a result, new collaborative performance measures are needed so that collaboration partners can make arrangements and compromises with each other, reflecting their common interests. In this paper, we first introduce the concept of collaborative key performance indicators (cKPIs), which are defined to measure the collaboration performance of multiple manufacturing partners. cKPIs are calculated by using several key performance indicators (KPIs) which individual partners can measure. For this research, we referred to the Supply Chain Operations Reference (SCOR) model (SCC, 2006) to define cKPI for manufacturing collaboration. Since the SCOR model provides corresponding performance metrics as well as several levels of supply chain process models, it can be a good reference for defining collaborative performance indicators (Barratt, 2004). In addition, we developed a two-folded desirability function to reflect the characteristics of performance indicators in manufacturing collaboration. The desirability function, which is based on the sigmoid function, can reflect multiple cKPI criteria in service level agreements (SLA). Further, unlike existing desirability functions, the sigmoid based desirability function can transform different scales of cKPIs into values between 0 and 1 without requiring maximum or minimum values (Lee and Yum, 2003). The weighted values of two-folded desirability functions for all cKPIs are summed to determine the synthetic performance of a collaboration, which can be compared with prior performance or partners performance. This paper is organized as follows. We first introduce the background of our research in Section 2. The framework of collaborative performance management is presented, along with the concept of cKPI, in Section 3. Subsequently, how to design the collaborative performance indicators and how to measure the performance indicators of manufacturing collaboration are described in Section 4 and Section 5, respectively. Finally, Section 6 concludes this paper. 2. BACKGROUND 2.1 Collaboration in Manufacturing Processes Manufacturing sector is a critical backbone of a nations economy while other industries such as information and service sectors are rapidly emerging for economic growth in developed countries. In order for manufacturing



A FRAMEWORK FOR THE ADOPTION OF RAPID PROTOTYPING FOR SMEs: FROM STRATEGIC TO OPERATIONAL

Ayyaz Ahmad*, Muhammad Ilyas Mazhar and Ian Howard Department of Mechanical Engineering,

Curtin University of Technology, WA 6102, Australia

*Corresponding author: Ayyaz Ahmad, [email protected] Rapidly changing global markets, unprecedented increase in product flexibility requirements and shorter product life cycles require more efficient technologies that can help reduce the time to market, which is considered to be a crucial factor to survive in todays highly volatile market conditions. Rapid prototyping technology (RPT) has the potential to make remarkable reductions in the product development time. However, its fast development pace combined with increasing complexity and variety has made the task of RPT selection difficult as well as challenging, resulting in low diffusion particularly at SME level. This paper systematically presents (i) Low RP adoption issues and challenges (ii) Importance of SMEs and the challenges they are facing to highlight the magnitude of the problem (iii) Previous work in the area of technology selection and adoption and finally offers an adoption framework which is exclusive for the adoption of RP technology by considering the manufacturing, operational, technology and cost drivers for a perfect technology fit into the business. Significance: Rapid Prototyping (RP) exhibits unique characteristics and can have potential impact on all business

functions, which demands a methodological approach for the evaluation and adoption of the technology. The main focus of this study is to propose a framework that facilitates the RP adoption from strategic to operational level to ensure complete and effective implementation to obtain the desired objectives, with a special emphasis on SMEs.

Keywords: Rapid prototyping, Technology adoption, SMEs, Technology Selection, Competitiveness

(Received 3 June 2011; Accepted in revised form 18 September 2011) 1. INTRODUCTION The changes in the global economic scenario have posed considerable threats to many companies, especially SMEs as they strive to stay competitive in world markets. This change in paradigms demands more flexibility in product designs. These challenges combined with increased variety and very short lead times has a great impact on the business of small to medium companies in securing a significant proportion of markets in which they operate. The conventional approaches and technologies are struggling to meet business needs. Consequently, manufacturers are searching for more efficient technologies, such as rapid prototyping that can help embrace the challenges. A critical activity for small companies is the decision-making on the selection and adoption of these advanced technologies. The SMEs task becomes more difficult because of the absence of any formal procedures (Ordoobadi et al., 2001). An advanced technology can be a great opportunity for a business but it can also be a threat to a company. A wrong alternative or too much investment in the right one can reduce the competitive advantage of a company (Trokkeli and Tuominen, 2002). The changing picture of the competition requires synchronization between business and new trends, which demands unique and effective solutions. These solutions should be designed to support them by keeping in view the specific nature of SMEs and ought to be simple, comprehensive and very practical so that they remain an effective part of the global value chain. To meet these global challenges, the design and manufacturing community is adopting the RP technology to remain efficient as well as competitive. The RP technology has enormous potential to shrink the product design and development timeline. Despite these great advantages, the adoption of RP at SMEs level is significantly low. A survey of 262 UK companies showed that 85% do not use RP. Lack of awareness of what the RP technology offers and how it can be successfully linked into the business functions are the key factors holding back this sector from the RP technology adoption. The majority of the groups who indicate that RP is irrelevant are unaware of what impact it can have on their business (Grenada, 2002). The condition is even worst in developing countries. Laar highlights the sensitivity of the issue by arguing that many engineers and R&D people are still unaware of the future implications of this technology. This is a major concern in view of the fact that technical departments are ignoring the RP/RM when it has already entered into world leading markets and has the potential to completely change the way we do business (Laar, 2007). Kidds argues that RP



AN INTEGRATED UTILISATION, SCHEDULING AND LOT-SIZING ALGORITHM FOR PULL PRODUCTION

Olufemi A.B. Adetunji, Venkata S.S. Yadavalli

Department of Industrial and Systems Engineering, University of Pretoria, Hatfield, Pretoria 0002, South Africa

We present an algorithm that continuously reduces the batch sizes of products on Non-constraining resource in a production network through the utilization of the idle time on such resource. This leads to reduction in the holding cost and increase in the frequency of batch release of the production system. This would also lead to reduction in customer facing supply lead time. Such technique could be valuable in typical pull production systems like lean manufacturing, theory of constraints or Constant-Work-in-Process CONWIP processes. An example is used to demonstrate a real life application of the algorithm, and it was found to work better for system cost minimization than a previous algorithm that uses the production run length as the criterion for batch reduction. Keywords: Lot-sizing, Utilization, Setup, Pull production, Scheduling algorithm

(Received 23 May 2011; Accepted in revised form 28 May 2012) 1. INTRODUCTION Traditionally, a lot size is taken to be the quantity of products contained in a production or purchase batch. This definition is also congruent to the classical batching model of economic order, which basically assumes that decision of what quantity to produce is made independently of job scheduling, but this is assumption is now being relaxed and the concept redefined. Potts and Wassenhove (1992), for instance, defined batching as making decision about whether or not to schedule similar jobs contiguously, and lot sizing as the decision about when and how to split the production of identical items into sub-lots. They noted that these decisions were traditionally taken as if lot sizing is independent of scheduling of jobs. This is obviated by the majority of the body of literature available on both subjects that are separate, with the impression being given that scheduling decisions are taken only after lot sizes of the various products have been decided. This assumption of independence is not usually true in most cases as the decisions are always inter-twined. Paul and Wassenhove also proposed a general model for integrated batching, lot sizing and scheduling. Drexl and Kims (1997) noted that lot-sizing and scheduling are two short term decisions of production planning that must be tied together with the medium term plan, which is the Master Production Scheduling of the system. Many models are since being published addressing integrated batching, lot sizing and scheduling. Potts and Kovalyov (2000) and Webster and Baker (1995) together with Potts and Wassenhove (1992) and Drexl and Kims (1997) are good readings. There is also a close relationship between system utilization and other system parameters like the Work-in-Process Inventory (WIP) and consequently the system holding cost and profitability. Variability in resource processing time and/or input arrival pattern have degrading influence on WIP level, especially as the system gets close to full utilization. This is succinctly summarized in Littles law. This effect of resource utilization on the production plan and the level of WIP appears not to have been well studied. Among the few known models incorporating resource utilization into production scheduling include Rappold and Yoho (2008), and a model proposed in Hopp (2008). The procedure proposed by Hopps is simple and straightforward to use, and that is what has been extended, and hopefully improved, in this paper. Next is a brief review of some work currently being done on integrated lot-sizing. We then proceed to briefly review some necessary principles of the management of constraint system pertinent to our model; especially the emphasis on balancing flow rather than capacities, which creates pockets of spare capacities (labor and machine), and the useful breakdown of the total cycle time of manufacturing resources and jobs, which identifies the various locations and quantities of idle capacities in the system, which can then be used in improved job scheduling due to reduced customer facing lead time and decreased lot sizes. The insight derived, however, is useful in other pull production environments as well since all pull techniques (including lean and CONWIP) always prefer to concentrate on flow and to buffer input and process variability via spare capacities as opposed to excess inventories. 2. INTEGRATED SCHEDULING AND LOT SIZING MODELS Solving integrated batching, lot sizing and scheduling problems has received more research attention recently. This could have also been buoyed by the development of many heuristics and techniques for solving difficult combinatorial problems. Among the recently published work in this area include Toledo et al (2010), which evaluated different parallel algorithms



THE OPTIMAL ORGANIZATION STRUCTURE DESIGN PROBLEM IN MAKE-TO-ORDER ENTERPRISES

Jess A. Mena

Department of Industrial Engineering, Monterrey Institute of Technology Campus, Chihuahua, Mexico

This paper addresses the organization structure design problem in a make-to-order (MTO) operation environment. A mathematical model is presented to aid an operations manager in an MTO environment to select a set of potential managerial layers to minimize the operation and supervision cost. With a given Work Breakdown Structure (WBS) for any specific project, solving this model leads an optimal organization structure design. The proposed model considers allocation tasks to workers, considering complexity and compatibility of each task with respect to workers, and the requirement of management for planning, execution, training and control in a hierarchical organization. This model addresses the span of control problem and provides a quantitative approach to the organization design problem and is intended for applications as a design tool in the make-to-order industries. Keywords Span of control, Organizational Design, Hierarchical Organization, Assignment Problem, Make-to-order

(Received 20 Sept 2011; Accepted in revised form 2 Jan 2012) 1. INTRODUCTION The span of management is perhaps the most discussed single concept in classical, neo-classical or modern management theory. Throughout its evolution it has been referred to by various titles such as span of management, span of control, span of supervision, and span of authority (Van Fleet & Benedian, 1977). The existing research work focus on principally qualitative methods to analyze this concept, i.e., heuristic rules based on experiences and/or intuition. This research develops an analytical modeling to determine the number of managerial layers and it is motivated in order to have an evaluation tool for functional based companies and also as a design tool for project-based companies. The challenge of mass customization brings great value to both the customer and the company. For example, building cars to customer order eliminates the need for companies to hold billions of dollars worth of finished stock. Any company able to free this capital would improve their competitive position, and be able to reinvest in future product development. The question for many company executives is how efficient the organizational structure could be. The need for frequent adjustment to an organizational structure can be found in this type of make-to-order or project-based companies, where work contents and its organizational structure could vary dramatically over a short period of time. This paper presents an analytical model for analyzing hierarchical organizations. It considers various factors that affect the requirement for supervision and formulates them into an analytical model which aims at optimizing the organizational design. This decision includes allocation tasks to workers, considering complexity and compatibility of each task with respect to workers, and the requirement of management for planning, execution, training and control in a hierarchical organization. The model is formulated as a 0-1 mixed integer program. The objective of the model is minimum operational cost, which are the sum of supervision costs at each level of the hierarchy and the number of workers assigned with tasks. This model addresses the span of control problem and provides a quantitative approach to the organization design problem and is intended for applications as a design tool in the make-to-order industries. Each project-based company may have to frequently readjust its organizational structure, as its capability and capacity shifts over time. It could also be applied to functionality based companies as an evaluation tool, to assess the optimality of their current organization structure. Meier and Bohte (Meier & Bohte, 2003) have recently reinvigorated the debate on span of control and the optimal manager-subordinate relationship. They offer a theory concerning the impacts and determinants of span of control and test it using data from educational organizations. The findings of Theobald et al. (Theobald & Nicholson-Crotty, S., 2005) suggest that manager-subordinates ratios, along with other structural influences on production, deserve considerably more attention than they have received in modern research on administration.



A NON-TRADITIONAL CAPITAL INVESTMENT CRITERIA-BASED METHOD TO OPTIMIZE A PORTFOLIO OF INVESTMENTS

Joana Siqueira de Souza1, Francisco Jos Kliemann Neto2, Michel Jos Anzanello3, Tiago Pascoal Filomena4

1Assistant Professor, Engineering School - Pontifcia Universidade Catolica of Rio Grande do Sul, Av. Ipiranga, 6681 - Partenon - 90619-900, Porto Alegre, RS, Brazil.

2Associate Professor, Department of Industrial and Transportation Engineering, Federal University of Rio Grande do Sul PPGEP/UFRGS. Av. Osvaldo Aranha, 99, 90035-190, Porto Alegre, RS, Brazil.

3Assistant Professor, Department of Industrial and Transportation Engineering, Federal University of Rio Grande do Sul PPGEP/UFRGS. Av. Osvaldo Aranha, 99, 90035-190, Porto Alegre, RS, Brazil.

4Assistant Professor, School Business, Federal University of Rio Grande do Sul Rua Washington Luiz, 855. Centro, 90010-460. Porto Alegre, RS, Brazil.

During the capital budgeting, companies need to define a set of projects that bring profitability, perpetuity and also have a direct link with the strategic objectives. This paper presents a practical model for defining a portfolio of industrial investments during capital budgeting by making use of traditional methods of investment analysis, such as Net Present Value (NPV), and by incorporating qualitative attributes on the analysis through the multicriteria analysis method called Non-Traditional Capital Investment Criteria (Boucher and MacStravic, 1991). Optimization techniques are then used to integrate economic and qualitative attributes subjected to budget restrictions. The proposed model was validated in an automotive company. Keywords: project portfolio, capital budgeting, net present value, multicriteria analysis, linear programming, decision-making.

(Received 31 Aug 2010; Accepted in revised form 1 Feb 2012)

1. INTRODUCTION The definition of a portfolio of projects in capital budgeting appears as an important issue in investment decisions and industrial planning (Chou et al. 2001). Decisions are seldom made for an isolated project; in most situations, the decision maker needs to consider several alternative projects relying on particular variables (Borgonovo and Peccati, 2006) associated not only to financial resources, but also to internal and external factors to the company (Kooros and Mcmanis, 1998; Mortensen et al. 2008). Although a large number of robust approaches related to investment decisions have been suggested in the literature, simplistic methods for evaluating investments are still widely used, and little structured decision making is applied in portfolio definition. Many assessment methods use discounted cash flow techniques such as the Internal Rate of Return (IRR), Net Present Value (NPV) and the Profitability Index (PI) (Cooper et al. 1997). More sophisticated methods can increase the likelihood of solid investments due to a stronger connection to company's strategy, leading to a more consistent analysis of opportunities (Verbeeten, 2006). Although many of these methods are appropriate for investment evaluation, Jansen et al. (2004) state they only enable tactical allocation of capital, and seldom take qualitative aspects into consideration (e.g. strategic aspects). That is corroborated by Arnold and Hatzopoulos (2000) who found that many firms invest their capital in non-economic projects (i.e. projects that do not necessarily bring economic benefits to the company), such as projects driven to workers health and safety. One way to incorporate qualitative aspects on decision-making process for capital investment is the adoption of multicriteria techniques, also known as Multiple Criteria Decision Making (MCDM) methods. A widespread method is the MAUT - Multiattribute Utility Theory - which relies on a simple and easy method for ranking the alternatives; see Min (1994). Another popular method is the Analytical Hierarchy Process (AHP), which hierarchically accommodates both quantitative and qualitative attributes of complex decisions (Saaty, 1980; Vaidya and Kumar, 2006). Successful applications of AHP can be found in Fogliatto and Guimares (2004), Rabbani et al. (2005), Vaidya and Kumar (2006), and Mendoza et al. (2008). A drawback of AHP is that it accommodates economic and qualitative aspects in different matrices, and also requires the comparison of all the alternatives over the same criteria. That is undesired when working with investment projects, since not all projects impact upon the same criteria. For example, a project to renew a truck fleet may have an impact on workers ergonomic condition, while a training project might not impact on that criterion. That led Boucher and MacStravic (1991) to develop an AHP-based multicriteria method for investment decision: the Non-Traditional Capital Investment Criteria (NC

recent-ijietap.pdf

Documents

new zealand smes

study of smes

new product launches

new products

web research

failed product

medium enterprises smes

funding research