Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

On the configuration and planning of dynamic manufacturing networks

On the configuration and planning of dynamic manufacturing networks Logist. Res. (2012) 5:105–111 DOI 10.1007/s12159-012-0086-9 OR IGINAL PAPER On the configuration and planning of dynamic manufacturing networks • • Nikolaos Papakostas Konstantinos Efthymiou Konstantinos Georgoulias George Chryssolouris Received: 3 August 2012 / Accepted: 23 August 2012 / Published online: 14 September 2012 Springer-Verlag 2012 Abstract Manufacturing organizations have been attempt- investigated toward optimization, in an attempt to improve ing to improve the operation of supply networks through product quality, to confront market competition, to shorten efficient supply chain management. Dynamic manufacturing lead times, as well as to reduce costs. These aspects con- networks (DMNs) constitute chains of diverse partners, whose stitute the main reason for the increasing complexity met in operation and interaction may change in a rapid and often not modern manufacturing systems. Controlling this com- predictable way. While the existing supply chain models plexity with conventional methods, such as the approaches are quite static and examine transportation modes, prod- based on manufacturing resource planning (MRP II) prin- uct changeover and production facility options with fixed ciples and concepts, requires more and more data and is suppliers and over a long period of time, the DMNs address becoming extremely difficult to manage. One of the top operations and risks on a daily basis. In this paper, a novel business pressures, dealt by enterprises, is the need to react decision-making approach is proposed for supporting the to demand changes in a timelier manner. Further to having process of configuring a DMN from a holistic perspective, to address the increase in year-over-year fulfillment and taking into account production, transportation and time con- transportation costs per unit, companies have been straints as well as multiple criteria such as time and cost. attempting to improve the cross-channel supply chain flexibility in order to achieve a faster reaction to demand Keywords Supply chain management  Scheduling  changes and to improve supply chain responsiveness [2]. Production planning  Logistics  Network design Manufacturing companies should be able to quickly restructure or transform the supply chain execution (source-deliver processes) in response to an evolving glo- 1 Introduction bal, multi-channel supply chain scenario. However, a lot of companies still do not have the ability to respond to dynamic demand cycles, while, at the same time, the In a volatile market environment, today’s manufacturing organizations strive to improve their performance, while increased globalization pushes the demand uncertainty at providing customers with more customization options [1]. even higher levels [2]. In the retail domain, for instance, The main classes of attributes to be considered when the demand has been so uncertain in the time span between making manufacturing decisions, that is, cost, time, quality mid 2010 and end of 2011 that the volume of inventory has and flexibility, are closely interrelated and have been either been too high or too low [2]. The recent events, concerning the volcano’s eruption in Iceland and the nuclear disaster in Fukushima, have reaffirmed the need for greater flexibility in order for manufacturing organizations N. Papakostas  K. Efthymiou  K. Georgoulias  to cope with the dynamic nature of the market and its G. Chryssolouris (&) fluctuations. Laboratory for Manufacturing Systems and Automation, At the same time, the existing, off-the-shelf Supply Department of Mechanical Engineering and Aeronautics, Chain Management software platforms and tools are too University of Patras, Patras, Greece expensive to be implemented and deployed at a broader e-mail: xrisol@lms.mech.upatras.gr 123 106 Logist. Res. (2012) 5:105–111 networked enterprise scale, including smaller companies Hierarchical approaches, initiated by the Original Equip- with limited Information and lower Communication ment Manufacturer (OEM), have also been proposed, capacity, and are unable to: where each partner’s tier performs all production planning activities and then provides these plans to the next tier for • Cover all actual phases of a manufacturing network carrying out its own process of production planning, until lifecycle and all tiers have completed their production planning activities • Cope with the highly dynamic and uncertain nature of [12]. Merging the planning activities of several partners demand. into one planning domain may improve the results of the It is not enough for today’s manufacturing enterprises to upstream collaboration [13]. Negotiation-based collabora- be networked: they have to be able to change and adapt to a tive planning approaches have been reported, focusing on continuously evolving environment and to form dynamic the use of upstream planning at the beginning and then on alliances with other companies and organizations in a fast the employment of a negotiation process in order for the and cost-efficient manner. overall performance to be improved [14]. The vast majority of the research work reported, dealing with the supply chain management and optimization, addres- 2 Current approaches for manufacturing sed very specific parts of the phases of a supply chain lifecycle. network management A few recent studies have dealt with the challenges related to each phase of the supply chain lifecycle in a more integrated The variations at trade barriers level and the worldwide evo- manner. The combined problem involving multiple transpor- lution of the transportation and communication means have tation modes, diverse supply chain flexibility options and led to the globalization of manufacturing activities [3]. New dynamic facility locations has been tackled in [8], experi- global strategies have pushed forward the internationalization menting with different adaptability schemes of a supply chain. of manufacturing systems [4]. The manufacturing landscape In [4], the integrated planning and transportation prob- has become more competitive, dynamic and complex. lem is addressed, proposing a mathematical model with A large number of studies have addressed various production and transportation capacity constraints. aspects of the supply chain management problem. The In general, so far, the approaches toward managing supply initial configuration of supply chains and the selection of chains have dealt with static instances of their operation: parts partners constitute one of the most critical phases in the or the entirety of the supply chain model are fixed and only a lifecycle of a supply network. A few research efforts have few alternative options are available. A few attempts deal with proposed the use of mixed-integer mathematical models different transportation modes, some others take into account with the objective to maximize profits or minimize the alternative facility locations and product changeover options overall supply chain operation costs [5]. Others have and very few, in principle the recent ones, propose a more focused on the identification of the optimum transportation sophisticated methodology in order for more facets of the modes for minimizing the total transportation and inven- problem to be addressed simultaneously. tory costs, including those addressing multi-product cases Our modeling approach allows for the formation of alter- for identifying optimal shipping times and loading policies native dynamic production network configurations as well as [6]. Production planning and transportation problems have for their validation via simulation in a series of network and also been addressed jointly [7]. Another stream of research demand settings, ensuring that the network be adaptive and work has dealt with the problem of having the supply chain capable of addressing the demand requirements. It may take flexibility increased, while retaining the capability to pro- into consideration partners who have not been part of the duce toward satisfying demand, by leveraging the alter- network in the past, requiring minimal information from their native supply chain options and the routing flexibility side regarding the initial configuration and planning of the within a pre-defined planning horizon [8]. The problem of manufacturing network. This way, a significant number of locating or relocating production facilities for satisfying the suppliers may be considered initially and therefore the chan- varying local demand has also been modeled by a few ces toward achieving an adaptive network configuration are researchers. In some cases, transportation mode and prod- significantly increased. At the same time, the uncertainty uct switching decisions have been addressed jointly [9, 10]. related to the demand, the production process and the trans- Collaborative planning of fixed supply networks is portation of products, subassemblies and parts may also be another issue that has attracted the interest of many considered, so that the risks regarding the operation of the research teams. The objective is to align the plans of the network be taken into account. individual supply chain partners and coordinate the pro- The development of highly adaptive manufacturing duction of the supply chain toward achieving a series of networks is a very important objective in today’s vola- common, or in some cases partner-specific, objectives [11]. tile environment. The proposed approach employs an 123 Logist. Res. (2012) 5:105–111 107 integrated holistic view of the network and attempts to • t: The time unit (e.g., day, shift, hour, etc.), t = 1…T, evaluate the performance of the network against multiple • T: The scheduling horizon, criteria, such as time and cost. At the same time, it offers a • A: The number of alternative DMN configurations to be mechanism for generating, evaluating and ranking a set of generated, • N: The number of samples (simulation runs) for each alternatives, so that the stakeholders involved be provided with more options, when having to decide about the con- alternative, • ED : The simulation end date of order o, figuration of a manufacturing network. the following information is required: 3 Dynamic manufacturing networks modeling • PP : This variable represents the bill of materials ij (BOM) of all products, subassemblies and parts that The manufacturing networks have to be more adaptive to may be produced or are available; when PP = 1, with ij the fluctuating demand in order for a more responsive and i = j, product i does not require other parts for being efficient operation to be achieved. Toward this direction, a produced. new modeling approach, employing a holistic view of the • SPC : The cost of manufacturing one unit of product sp overall network performance, is proposed. The major steps p in partner s. are depicted in Fig. 1. • SPI : The inventory cost per unit of product p in the sp The principle objective is to use minimal information, facilities of partner s. so that potential partners with minimal Information and • SSR : The cost of transferring one unit of product ss pm Communication capacity may take part in a dynamic p from partner s to partner s using transportation mode manufacturing network (DMN). m. • SST : The time required for transferring one unit of ss pm 3.1 Information requirements product p from partner s to partner s , using transpor- tation mode m. This approach requires that some minimal information • SSTV 0 : The stochastic variation of the time required ss pm regarding the production orders and the partners’ capacity for transferring one unit of product p from partner s to and network be available, in order for different alternative partner s , using transportation mode m, following a DMN configurations to be generated and evaluated. 0 0 uniform distribution [-SSTV , SSTV ]. ss pm ss pm Assuming that: • SP : The capacity per time unit required for producing sp • S: The overall number of partners (including manufac- product p in the facilities of partner s, with 0 B SP B 1, s = 1…S, p = 1…P. turers, suppliers and customers), sp • P: The number of products, subassemblies and parts, • SPV : The stochastic variation of capacity per time sp unit required for producing product p in the facilities of • O: The overall number of orders, partner s, following a uniform distribution [-SPV , • M: The number of different modes of transportation sp SPV ]. (e.g., ground, air, etc.), sp • S : the maximum number of partners that may max produce the same part within the DMN. Manufacturing, Generation of Alternative DMN Transportation • ST : The capacity already allocated in time unit t for configurations st Constraints partner s. • STV : The stochastic variation, regarding the capacity st already allocated in time unit t for partner s, following a Production, Demand Simulation of Alternatives Samples uniform distribution [-STV , STV ]. st st Uncertainty • SY : The quantity of product p in the inventory of sp partner s. • PO : The quantity of product p of order o, issued by ops Performance Evaluation of Alternatives partner s. Criteria • DD : The due date of order o. • AD : The arrival date of order o. Criteria The above represent the information required for gen- Best Alternatives Weights erating alternative DMN configurations, without having to take into account the process plans and the specific details Fig. 1 Overview of the proposed approach of each partner’s production equipment. 123 108 Logist. Res. (2012) 5:105–111 Table 1 An alternative DMN configuration example Assignment of Orders to DMN Partners Orders Arrival based on matrix A sp Partner Product P Product P Product P Product P 1 2 3 4 S 0.0 0.4 0.0 0.0 S 1.0 0.6 0.0 0.0 Checking Assigned Each DMN Partner sends existing quantities, Orders and S 0.0 0.0 0.8 0.0 selecting a transportation mode Inventory S 0.0 0.0 0.0 0.7 S 0.0 0.0 0.2 0.3 S 0.0 0.0 0.0 0.0 Each DMN Partner estimates the quantities S 0.0 0.0 0.0 0.0 that may be produced and plans production 3.2 Generation of alternative DMN configurations Each DMN Partner releases orders for extra parts or subassemblies We define as an alternative DMN configuration the SxP matrix A , where each element of this matrix a repre- sp sp sents the probability that partner s produces product p. Each DMN partner checks rest of orders and This probability actually defines which partner will be updates the inventory when production ends producing which product, part or subassembly, when an order (either for an end product or for a subassembly or a Fig. 2 Overview of the simulation process part required for manufacturing the end product) arrives or is issued within the DMN. An example of an alternative DMN configuration transportation efficiency, are, a random transportation (matrix A ) is shown in Table 1: with reference to the case sp mode m from the ones available is selected for each sam- scenario described in Sect. 4 (alternative #4 of Table 4), ple. The associated transportation cost and time SSR , ss pm where 5 suppliers (S –S ) and 2 customers (S and S ) have 1 5 6 7 0 0 SST , SSTV are used in the process of calculating ss pm ss pm to collaborate for the dispatch of a number of orders, the corresponding transportation cost and time of order product P will entirely be produced by S , while partner S 1 2 1 o for sample n (TC ). The remaining product quantities of on will produce 40 % of the quantity ordered of P and S will 2 2 the assigned orders are then checked against their produce the remaining 60 % of the quantity ordered of P . requirements of subassemblies and parts. If the production We consider as a DMN the set of all potential partners that for a part of the order may be initiated, a production order could take part in the dispatching of an order. Contrary to is released and planned, having taken into account the the existing hierarchical approaches, the cooperation production capacity already allocated (ST , STV ) as well st st among the DMN members is considered being loose, as the capacity requirements of the products to be produced without having to identify which partners have a leading (SP , SPV ). In case extra subassemblies or parts are sp sp role or not. Orders may actually be received by all partners. required for the fulfillment of an order, new ones are In this paper, however, it is assumed that the partners who released toward the DMN partners. When all orders have can manufacture and deliver a specific product are the ones been dispatched, the simulation of the samples is com- who usually receive an order for this product and therefore pleted and other ones are then simulated until all N samples initiate the DMN configuration process. of all A alternatives are evaluated (Fig. 3). 3.3 Simulation of alternatives samples 3.4 Evaluation of alternatives For each alternative DMN configuration, a number of All the samples of alternatives are evaluated against the samples is simulated (Fig. 2). For each sample, in each criteria of average tardiness and cost. In particular, time unit, the orders received are randomly assigned to the P P N 0 partners available, the ones who can produce the products fmaxðED  DD Þ; 0g on on m¼1 o¼1 Tard ¼ ð1Þ ordered, as per the matrix A . Each partner checks the sp assigned orders and in case a part of an order may be P P N 0 TC on n¼1 o¼1 fulfilled, a transfer order is released toward the partner who Cost ¼ ð2Þ has released the original order. In order to take into account different transportation options in all samples, thus con- Using the simple additive weight method and having sidering how adaptive the DMN configurations, in terms of already identified the criteria weights for defining their 123 Logist. Res. (2012) 5:105–111 109 Fig. 3 An example with Alternatives Samples alternatives and samples Per sample, stochastic: � Available production capacity � Average Cost for a a a 1 12 � Production capacity required � Average Tardiness for a � Transportation time � Transportation mode 1n Per sample, stochastic: � Available production capacity � Average Cost for a a a 2 22 � Production capacity required � Average Tardiness for a � Transportation time � Transportation mode 2n relative importance, the overall utility of each alternative Product P may be calculated with the aid of a software application. (Partners S or S ) 1 2 This way, all alternatives may then be ranked and pre- sented to the user. The average cost and tardiness values of the alternative DMN configurations are considered as a Product P Product P 3 4 measure of the DMN’s adaptability toward demand (Partners S or S or S ) (Partners S or S or S ) 3 4 5 3 4 5 requirements. Apparently, future demand scenarios may also be taken into consideration for each alternative. Product P (Partner S ) 4 Implementation and experiments Fig. 4 Case scenario: bill of materials and suitable partners For the purpose of testing and validating this proposed approach, a software application with a simulation engine has been implemented and a series of experiments have been carried out. A 3-tier case scenario is demonstrated Table 2 Description of the case scenario with 7 partners (including 2 customers) and 4 products. DMN properties Value Part P may be produced by partner S and S , whereas P 1 1 2 2 is produced by S only and P and P may be produced by 2 3 4 Number of partners 7 partners S , S , S (Fig. 4). 3 4 5 Number of products 4 The properties of the DMN are shown in Table 2. Number of tiers 3 The information regarding the orders is depicted in Table 3. Transportation modes 2 Four experiments have been carried out with a different Evaluation criteria and weights Cost: 50 %, tardiness: 50 % number of alternatives (A) and a maximum number of 123 110 Logist. Res. (2012) 5:105–111 partners (S ) who could take part in the manufacturing of • The uncertainty associated with the production and max the same product or part. For the first two experiments only transportation times, as well as with the demand profile, one partner may produce each part, while in experiments 3 is also considered via the sampling mechanism of and 4, up to 2 partners may produce each part. The results the proposed approach: many different scenarios are therefore simulated beforehand, in order to ensure that of the best alternative generated in each experiment are shown in Table 4. the manufacturing network may operate efficiently under different conditions. Twenty samples were generated per alternative for all four experiments. The performance of the best alternatives • This method enables collaboration schemes of specific products, subassemblies and parts, that is, their pro- suggested in these experiments is compared and their utility is estimated, taking into account the criteria weights. duction may be distributed to many partners. The It is obvious that the more alternatives are generated, uncertainty related to the partners’ production capacity simulated and evaluated, the more promising the best is taken into consideration and therefore collaborative alternative DMN configuration looks. It is also interesting schemes with more partners are proposed in case it is to note that the performance of the DMN is better when likely that a partner cannot deliver. more options are available, in terms of the maximum • The different transportation modes provided are also taken into account, along with the corresponding costs number of partners that can produce the same part. and times for each alternative via the sampling mechanism. This way, the adaptability of the proposed 5 Conclusions DMN configurations in terms of how well they behave in terms of transportation efficiency is considered; in A novel approach for modeling DMN as well as for gen- case any transportation problems emerge, the proposed erating and evaluating alternative configurations has been DMN configurations are expected to cope well with proposed. This method requires minimal information these problems. regarding the status of the manufacturing systems belong- • Whenever a disruption in the operation of a DMN occurs, the proposed approach may be executed again, ing to the network partners. This information is in principle limited to the capacity available per partner over the toward modifying the initial DMN configuration. scheduling horizon, their production capabilities, the status Nevertheless, a series of assumptions were made for of their inventory and the existing modes of transportation. testing, validating and presenting the proposed approach: The dynamic nature of the manufacturing network is addressed in the following ways: • Production capacities have been assumed to be evenly distributed, • A randomly generated demand profile was used Table 3 Orders information including the orders’ due dates. Order # Product Customer Quantity Due date (days) However, without loss of generality, the proposed methodology may easily be used with other statistical 1 P S 12 1 6 distributions and demand profiles. 2 P S 24 1 7 Through the simple case scenario given and the exper- 3 P S 27 1 6 iments carried out, it has been shown that the proposed 4 P S 12 2 7 approach could be used for determining adaptive DMNs in 5 P S 23 3 7 a volatile and highly uncertain global market environment. Table 4 Experiments and performance of best alternatives # AS Cost (€) Tard (days) Util P P P P max a a 1 2 3 4 partners partners partners partners 1 5 1 55,400 5.24 0.00 S (100 %) S (100 %) S (100 %) S (100 %) 2 1 3 4 2 50 1 36,765 4.20 0.88 S (100 %) S (100 %) S (100 %) S (100 %) 1 2 3 4 3 5 2 41,747 4.53 0.63 S (40 %) S (70 %) S (40 %) S (50 %) 1 1 3 3 S (60 %) S (30 %) S (60 %) S (50 %) 2 2 5 4 4 50 2 38,275 3.87 0.96 S (100 %) S (40 %) S (80 %) S (70 %) 2 1 3 4 S (60 %) S (20 %) S (30 %) 2 5 5 123 Logist. Res. (2012) 5:105–111 111 2. Permenter K, Anand S (2011) State of cross channel retail supply The problem of integrating complex products/parts and chain execution. Aberdeen Group, Boston suppliers’ interrelationships, the finite production capacity 3. Chryssolouris G, Papakostas N, Mavrikios D (2008) A perspec- of the potential partners, different transportation modes and tive on manufacturing strategy: produce more with less. CIRP J the uncertainty pertaining to available and required pro- Manuf Sci Technol 1:45–52 4. Scholz-Reiter B, Morosini Frazzon E, Makuschewitz T (2010) duction capacities and process times cannot be handled by Integrating manufacturing and logistic systems along global conventional Mathematical Programming and Operations supply chains. CIRP J Manuf Sci Technol 2:216–223 Research approaches. 5. Viswanadham N, Gaonkar RS (2003) Partners selection and Going beyond the configuration and planning phases, synchronized planning in dynamic manufacturing networks. IEEE Transact Robot Autom 19:117–130 further features would include options for lot sizing within 6. Speranza MG, Ukovich W (1994) Minimizing transportation and the DMN as well as options for expanding the use of the inventory costs for several products on a single link. Oper Res proposed approach in the domain of the manufacturing 42:879–894 scheduling, where detailed process plans and configura- 7. Theattre K, Graves S (2001) Tactical shipping and scheduling at Polaroid with dual lead-times. In: Innovation in manufacturing tions have to be considered at each partner’s level. Inte- systems and technology (IMST) report, http://web.mit.edu/ grating data from the shop floor and the logistics network sgraves/www/SMA%20paper%20Oct%2001.pdf. Accessed 3 for monitoring the operation of a DMN is also another idea August 2012 that is worth experimenting with. More sophisticated sce- 8. Seifert RW, Langenberg KU (2011) Managing business dynamics with adaptive supply chain portfolios. Eur J Oper Res 215:551– narios may also be tested, involving the transportation activities and organizations as part of the DMN. 9. Vidal CJ, Goetschalckx M (2001) A global supply chain model DMNs are expected to be in charge of an increasing part with transfer pricing and transportation cost allocation. Eur J of the global manufacturing activity and therefore provid- Oper Res 129:134–158 10. Klincewicz JG, Luss H, Yu CS (1988) A large-scale multiloca- ing new methods and tools for improving their operation, tion capacity planning model. Eur J Oper Res 34:178–190 and overall efficiency is of paramount importance. 11. Stadtler H (2009) A framework for collaborative planning and state-of-the-art. OR Spectrum 31:5–30 Acknowledgments The work in this paper has been partially sup- 12. Bhatnagar R, Chandra P, Goyal SK (2004) Models for multi-plant ported by the FP7 Integrated Project ‘‘IMAGINE—Innovative end-to- coordination. Eur J Oper Res 67:141–160 end Management of Dynamic Manufacturing Networks,’’ funded by 13. Pibernik R, Sucky R (2007) An approach to inter domain master the CEU. planning in supply chains. Int J Prod Econ 108:200–221 14. Dudek G, Stadtler H (2005) Negotiation-based collaborative planning between supply chain partners. Eur J Oper Res 163: 668–687 References 1. Chryssolouris G (2006) Manufacturing systems—theory and practice. Springer, New York http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Logistics Research Springer Journals

On the configuration and planning of dynamic manufacturing networks

Download PDF
7 pages

Loading next page...
 
/lp/springer-journals/on-the-configuration-and-planning-of-dynamic-manufacturing-networks-v4U0EO9o4S

References (15)

Publisher
Springer Journals
Copyright
Copyright © 2012 by Springer-Verlag
Subject
Engineering; Engineering Economics, Organization, Logistics, Marketing; Logistics; Industrial and Production Engineering; Simulation and Modeling; Operation Research/Decision Theory
ISSN
1865-035X
eISSN
1865-0368
DOI
10.1007/s12159-012-0086-9
Publisher site
See Article on Publisher Site

Abstract

Logist. Res. (2012) 5:105–111 DOI 10.1007/s12159-012-0086-9 OR IGINAL PAPER On the configuration and planning of dynamic manufacturing networks • • Nikolaos Papakostas Konstantinos Efthymiou Konstantinos Georgoulias George Chryssolouris Received: 3 August 2012 / Accepted: 23 August 2012 / Published online: 14 September 2012 Springer-Verlag 2012 Abstract Manufacturing organizations have been attempt- investigated toward optimization, in an attempt to improve ing to improve the operation of supply networks through product quality, to confront market competition, to shorten efficient supply chain management. Dynamic manufacturing lead times, as well as to reduce costs. These aspects con- networks (DMNs) constitute chains of diverse partners, whose stitute the main reason for the increasing complexity met in operation and interaction may change in a rapid and often not modern manufacturing systems. Controlling this com- predictable way. While the existing supply chain models plexity with conventional methods, such as the approaches are quite static and examine transportation modes, prod- based on manufacturing resource planning (MRP II) prin- uct changeover and production facility options with fixed ciples and concepts, requires more and more data and is suppliers and over a long period of time, the DMNs address becoming extremely difficult to manage. One of the top operations and risks on a daily basis. In this paper, a novel business pressures, dealt by enterprises, is the need to react decision-making approach is proposed for supporting the to demand changes in a timelier manner. Further to having process of configuring a DMN from a holistic perspective, to address the increase in year-over-year fulfillment and taking into account production, transportation and time con- transportation costs per unit, companies have been straints as well as multiple criteria such as time and cost. attempting to improve the cross-channel supply chain flexibility in order to achieve a faster reaction to demand Keywords Supply chain management  Scheduling  changes and to improve supply chain responsiveness [2]. Production planning  Logistics  Network design Manufacturing companies should be able to quickly restructure or transform the supply chain execution (source-deliver processes) in response to an evolving glo- 1 Introduction bal, multi-channel supply chain scenario. However, a lot of companies still do not have the ability to respond to dynamic demand cycles, while, at the same time, the In a volatile market environment, today’s manufacturing organizations strive to improve their performance, while increased globalization pushes the demand uncertainty at providing customers with more customization options [1]. even higher levels [2]. In the retail domain, for instance, The main classes of attributes to be considered when the demand has been so uncertain in the time span between making manufacturing decisions, that is, cost, time, quality mid 2010 and end of 2011 that the volume of inventory has and flexibility, are closely interrelated and have been either been too high or too low [2]. The recent events, concerning the volcano’s eruption in Iceland and the nuclear disaster in Fukushima, have reaffirmed the need for greater flexibility in order for manufacturing organizations N. Papakostas  K. Efthymiou  K. Georgoulias  to cope with the dynamic nature of the market and its G. Chryssolouris (&) fluctuations. Laboratory for Manufacturing Systems and Automation, At the same time, the existing, off-the-shelf Supply Department of Mechanical Engineering and Aeronautics, Chain Management software platforms and tools are too University of Patras, Patras, Greece expensive to be implemented and deployed at a broader e-mail: xrisol@lms.mech.upatras.gr 123 106 Logist. Res. (2012) 5:105–111 networked enterprise scale, including smaller companies Hierarchical approaches, initiated by the Original Equip- with limited Information and lower Communication ment Manufacturer (OEM), have also been proposed, capacity, and are unable to: where each partner’s tier performs all production planning activities and then provides these plans to the next tier for • Cover all actual phases of a manufacturing network carrying out its own process of production planning, until lifecycle and all tiers have completed their production planning activities • Cope with the highly dynamic and uncertain nature of [12]. Merging the planning activities of several partners demand. into one planning domain may improve the results of the It is not enough for today’s manufacturing enterprises to upstream collaboration [13]. Negotiation-based collabora- be networked: they have to be able to change and adapt to a tive planning approaches have been reported, focusing on continuously evolving environment and to form dynamic the use of upstream planning at the beginning and then on alliances with other companies and organizations in a fast the employment of a negotiation process in order for the and cost-efficient manner. overall performance to be improved [14]. The vast majority of the research work reported, dealing with the supply chain management and optimization, addres- 2 Current approaches for manufacturing sed very specific parts of the phases of a supply chain lifecycle. network management A few recent studies have dealt with the challenges related to each phase of the supply chain lifecycle in a more integrated The variations at trade barriers level and the worldwide evo- manner. The combined problem involving multiple transpor- lution of the transportation and communication means have tation modes, diverse supply chain flexibility options and led to the globalization of manufacturing activities [3]. New dynamic facility locations has been tackled in [8], experi- global strategies have pushed forward the internationalization menting with different adaptability schemes of a supply chain. of manufacturing systems [4]. The manufacturing landscape In [4], the integrated planning and transportation prob- has become more competitive, dynamic and complex. lem is addressed, proposing a mathematical model with A large number of studies have addressed various production and transportation capacity constraints. aspects of the supply chain management problem. The In general, so far, the approaches toward managing supply initial configuration of supply chains and the selection of chains have dealt with static instances of their operation: parts partners constitute one of the most critical phases in the or the entirety of the supply chain model are fixed and only a lifecycle of a supply network. A few research efforts have few alternative options are available. A few attempts deal with proposed the use of mixed-integer mathematical models different transportation modes, some others take into account with the objective to maximize profits or minimize the alternative facility locations and product changeover options overall supply chain operation costs [5]. Others have and very few, in principle the recent ones, propose a more focused on the identification of the optimum transportation sophisticated methodology in order for more facets of the modes for minimizing the total transportation and inven- problem to be addressed simultaneously. tory costs, including those addressing multi-product cases Our modeling approach allows for the formation of alter- for identifying optimal shipping times and loading policies native dynamic production network configurations as well as [6]. Production planning and transportation problems have for their validation via simulation in a series of network and also been addressed jointly [7]. Another stream of research demand settings, ensuring that the network be adaptive and work has dealt with the problem of having the supply chain capable of addressing the demand requirements. It may take flexibility increased, while retaining the capability to pro- into consideration partners who have not been part of the duce toward satisfying demand, by leveraging the alter- network in the past, requiring minimal information from their native supply chain options and the routing flexibility side regarding the initial configuration and planning of the within a pre-defined planning horizon [8]. The problem of manufacturing network. This way, a significant number of locating or relocating production facilities for satisfying the suppliers may be considered initially and therefore the chan- varying local demand has also been modeled by a few ces toward achieving an adaptive network configuration are researchers. In some cases, transportation mode and prod- significantly increased. At the same time, the uncertainty uct switching decisions have been addressed jointly [9, 10]. related to the demand, the production process and the trans- Collaborative planning of fixed supply networks is portation of products, subassemblies and parts may also be another issue that has attracted the interest of many considered, so that the risks regarding the operation of the research teams. The objective is to align the plans of the network be taken into account. individual supply chain partners and coordinate the pro- The development of highly adaptive manufacturing duction of the supply chain toward achieving a series of networks is a very important objective in today’s vola- common, or in some cases partner-specific, objectives [11]. tile environment. The proposed approach employs an 123 Logist. Res. (2012) 5:105–111 107 integrated holistic view of the network and attempts to • t: The time unit (e.g., day, shift, hour, etc.), t = 1…T, evaluate the performance of the network against multiple • T: The scheduling horizon, criteria, such as time and cost. At the same time, it offers a • A: The number of alternative DMN configurations to be mechanism for generating, evaluating and ranking a set of generated, • N: The number of samples (simulation runs) for each alternatives, so that the stakeholders involved be provided with more options, when having to decide about the con- alternative, • ED : The simulation end date of order o, figuration of a manufacturing network. the following information is required: 3 Dynamic manufacturing networks modeling • PP : This variable represents the bill of materials ij (BOM) of all products, subassemblies and parts that The manufacturing networks have to be more adaptive to may be produced or are available; when PP = 1, with ij the fluctuating demand in order for a more responsive and i = j, product i does not require other parts for being efficient operation to be achieved. Toward this direction, a produced. new modeling approach, employing a holistic view of the • SPC : The cost of manufacturing one unit of product sp overall network performance, is proposed. The major steps p in partner s. are depicted in Fig. 1. • SPI : The inventory cost per unit of product p in the sp The principle objective is to use minimal information, facilities of partner s. so that potential partners with minimal Information and • SSR : The cost of transferring one unit of product ss pm Communication capacity may take part in a dynamic p from partner s to partner s using transportation mode manufacturing network (DMN). m. • SST : The time required for transferring one unit of ss pm 3.1 Information requirements product p from partner s to partner s , using transpor- tation mode m. This approach requires that some minimal information • SSTV 0 : The stochastic variation of the time required ss pm regarding the production orders and the partners’ capacity for transferring one unit of product p from partner s to and network be available, in order for different alternative partner s , using transportation mode m, following a DMN configurations to be generated and evaluated. 0 0 uniform distribution [-SSTV , SSTV ]. ss pm ss pm Assuming that: • SP : The capacity per time unit required for producing sp • S: The overall number of partners (including manufac- product p in the facilities of partner s, with 0 B SP B 1, s = 1…S, p = 1…P. turers, suppliers and customers), sp • P: The number of products, subassemblies and parts, • SPV : The stochastic variation of capacity per time sp unit required for producing product p in the facilities of • O: The overall number of orders, partner s, following a uniform distribution [-SPV , • M: The number of different modes of transportation sp SPV ]. (e.g., ground, air, etc.), sp • S : the maximum number of partners that may max produce the same part within the DMN. Manufacturing, Generation of Alternative DMN Transportation • ST : The capacity already allocated in time unit t for configurations st Constraints partner s. • STV : The stochastic variation, regarding the capacity st already allocated in time unit t for partner s, following a Production, Demand Simulation of Alternatives Samples uniform distribution [-STV , STV ]. st st Uncertainty • SY : The quantity of product p in the inventory of sp partner s. • PO : The quantity of product p of order o, issued by ops Performance Evaluation of Alternatives partner s. Criteria • DD : The due date of order o. • AD : The arrival date of order o. Criteria The above represent the information required for gen- Best Alternatives Weights erating alternative DMN configurations, without having to take into account the process plans and the specific details Fig. 1 Overview of the proposed approach of each partner’s production equipment. 123 108 Logist. Res. (2012) 5:105–111 Table 1 An alternative DMN configuration example Assignment of Orders to DMN Partners Orders Arrival based on matrix A sp Partner Product P Product P Product P Product P 1 2 3 4 S 0.0 0.4 0.0 0.0 S 1.0 0.6 0.0 0.0 Checking Assigned Each DMN Partner sends existing quantities, Orders and S 0.0 0.0 0.8 0.0 selecting a transportation mode Inventory S 0.0 0.0 0.0 0.7 S 0.0 0.0 0.2 0.3 S 0.0 0.0 0.0 0.0 Each DMN Partner estimates the quantities S 0.0 0.0 0.0 0.0 that may be produced and plans production 3.2 Generation of alternative DMN configurations Each DMN Partner releases orders for extra parts or subassemblies We define as an alternative DMN configuration the SxP matrix A , where each element of this matrix a repre- sp sp sents the probability that partner s produces product p. Each DMN partner checks rest of orders and This probability actually defines which partner will be updates the inventory when production ends producing which product, part or subassembly, when an order (either for an end product or for a subassembly or a Fig. 2 Overview of the simulation process part required for manufacturing the end product) arrives or is issued within the DMN. An example of an alternative DMN configuration transportation efficiency, are, a random transportation (matrix A ) is shown in Table 1: with reference to the case sp mode m from the ones available is selected for each sam- scenario described in Sect. 4 (alternative #4 of Table 4), ple. The associated transportation cost and time SSR , ss pm where 5 suppliers (S –S ) and 2 customers (S and S ) have 1 5 6 7 0 0 SST , SSTV are used in the process of calculating ss pm ss pm to collaborate for the dispatch of a number of orders, the corresponding transportation cost and time of order product P will entirely be produced by S , while partner S 1 2 1 o for sample n (TC ). The remaining product quantities of on will produce 40 % of the quantity ordered of P and S will 2 2 the assigned orders are then checked against their produce the remaining 60 % of the quantity ordered of P . requirements of subassemblies and parts. If the production We consider as a DMN the set of all potential partners that for a part of the order may be initiated, a production order could take part in the dispatching of an order. Contrary to is released and planned, having taken into account the the existing hierarchical approaches, the cooperation production capacity already allocated (ST , STV ) as well st st among the DMN members is considered being loose, as the capacity requirements of the products to be produced without having to identify which partners have a leading (SP , SPV ). In case extra subassemblies or parts are sp sp role or not. Orders may actually be received by all partners. required for the fulfillment of an order, new ones are In this paper, however, it is assumed that the partners who released toward the DMN partners. When all orders have can manufacture and deliver a specific product are the ones been dispatched, the simulation of the samples is com- who usually receive an order for this product and therefore pleted and other ones are then simulated until all N samples initiate the DMN configuration process. of all A alternatives are evaluated (Fig. 3). 3.3 Simulation of alternatives samples 3.4 Evaluation of alternatives For each alternative DMN configuration, a number of All the samples of alternatives are evaluated against the samples is simulated (Fig. 2). For each sample, in each criteria of average tardiness and cost. In particular, time unit, the orders received are randomly assigned to the P P N 0 partners available, the ones who can produce the products fmaxðED  DD Þ; 0g on on m¼1 o¼1 Tard ¼ ð1Þ ordered, as per the matrix A . Each partner checks the sp assigned orders and in case a part of an order may be P P N 0 TC on n¼1 o¼1 fulfilled, a transfer order is released toward the partner who Cost ¼ ð2Þ has released the original order. In order to take into account different transportation options in all samples, thus con- Using the simple additive weight method and having sidering how adaptive the DMN configurations, in terms of already identified the criteria weights for defining their 123 Logist. Res. (2012) 5:105–111 109 Fig. 3 An example with Alternatives Samples alternatives and samples Per sample, stochastic: � Available production capacity � Average Cost for a a a 1 12 � Production capacity required � Average Tardiness for a � Transportation time � Transportation mode 1n Per sample, stochastic: � Available production capacity � Average Cost for a a a 2 22 � Production capacity required � Average Tardiness for a � Transportation time � Transportation mode 2n relative importance, the overall utility of each alternative Product P may be calculated with the aid of a software application. (Partners S or S ) 1 2 This way, all alternatives may then be ranked and pre- sented to the user. The average cost and tardiness values of the alternative DMN configurations are considered as a Product P Product P 3 4 measure of the DMN’s adaptability toward demand (Partners S or S or S ) (Partners S or S or S ) 3 4 5 3 4 5 requirements. Apparently, future demand scenarios may also be taken into consideration for each alternative. Product P (Partner S ) 4 Implementation and experiments Fig. 4 Case scenario: bill of materials and suitable partners For the purpose of testing and validating this proposed approach, a software application with a simulation engine has been implemented and a series of experiments have been carried out. A 3-tier case scenario is demonstrated Table 2 Description of the case scenario with 7 partners (including 2 customers) and 4 products. DMN properties Value Part P may be produced by partner S and S , whereas P 1 1 2 2 is produced by S only and P and P may be produced by 2 3 4 Number of partners 7 partners S , S , S (Fig. 4). 3 4 5 Number of products 4 The properties of the DMN are shown in Table 2. Number of tiers 3 The information regarding the orders is depicted in Table 3. Transportation modes 2 Four experiments have been carried out with a different Evaluation criteria and weights Cost: 50 %, tardiness: 50 % number of alternatives (A) and a maximum number of 123 110 Logist. Res. (2012) 5:105–111 partners (S ) who could take part in the manufacturing of • The uncertainty associated with the production and max the same product or part. For the first two experiments only transportation times, as well as with the demand profile, one partner may produce each part, while in experiments 3 is also considered via the sampling mechanism of and 4, up to 2 partners may produce each part. The results the proposed approach: many different scenarios are therefore simulated beforehand, in order to ensure that of the best alternative generated in each experiment are shown in Table 4. the manufacturing network may operate efficiently under different conditions. Twenty samples were generated per alternative for all four experiments. The performance of the best alternatives • This method enables collaboration schemes of specific products, subassemblies and parts, that is, their pro- suggested in these experiments is compared and their utility is estimated, taking into account the criteria weights. duction may be distributed to many partners. The It is obvious that the more alternatives are generated, uncertainty related to the partners’ production capacity simulated and evaluated, the more promising the best is taken into consideration and therefore collaborative alternative DMN configuration looks. It is also interesting schemes with more partners are proposed in case it is to note that the performance of the DMN is better when likely that a partner cannot deliver. more options are available, in terms of the maximum • The different transportation modes provided are also taken into account, along with the corresponding costs number of partners that can produce the same part. and times for each alternative via the sampling mechanism. This way, the adaptability of the proposed 5 Conclusions DMN configurations in terms of how well they behave in terms of transportation efficiency is considered; in A novel approach for modeling DMN as well as for gen- case any transportation problems emerge, the proposed erating and evaluating alternative configurations has been DMN configurations are expected to cope well with proposed. This method requires minimal information these problems. regarding the status of the manufacturing systems belong- • Whenever a disruption in the operation of a DMN occurs, the proposed approach may be executed again, ing to the network partners. This information is in principle limited to the capacity available per partner over the toward modifying the initial DMN configuration. scheduling horizon, their production capabilities, the status Nevertheless, a series of assumptions were made for of their inventory and the existing modes of transportation. testing, validating and presenting the proposed approach: The dynamic nature of the manufacturing network is addressed in the following ways: • Production capacities have been assumed to be evenly distributed, • A randomly generated demand profile was used Table 3 Orders information including the orders’ due dates. Order # Product Customer Quantity Due date (days) However, without loss of generality, the proposed methodology may easily be used with other statistical 1 P S 12 1 6 distributions and demand profiles. 2 P S 24 1 7 Through the simple case scenario given and the exper- 3 P S 27 1 6 iments carried out, it has been shown that the proposed 4 P S 12 2 7 approach could be used for determining adaptive DMNs in 5 P S 23 3 7 a volatile and highly uncertain global market environment. Table 4 Experiments and performance of best alternatives # AS Cost (€) Tard (days) Util P P P P max a a 1 2 3 4 partners partners partners partners 1 5 1 55,400 5.24 0.00 S (100 %) S (100 %) S (100 %) S (100 %) 2 1 3 4 2 50 1 36,765 4.20 0.88 S (100 %) S (100 %) S (100 %) S (100 %) 1 2 3 4 3 5 2 41,747 4.53 0.63 S (40 %) S (70 %) S (40 %) S (50 %) 1 1 3 3 S (60 %) S (30 %) S (60 %) S (50 %) 2 2 5 4 4 50 2 38,275 3.87 0.96 S (100 %) S (40 %) S (80 %) S (70 %) 2 1 3 4 S (60 %) S (20 %) S (30 %) 2 5 5 123 Logist. Res. (2012) 5:105–111 111 2. Permenter K, Anand S (2011) State of cross channel retail supply The problem of integrating complex products/parts and chain execution. Aberdeen Group, Boston suppliers’ interrelationships, the finite production capacity 3. Chryssolouris G, Papakostas N, Mavrikios D (2008) A perspec- of the potential partners, different transportation modes and tive on manufacturing strategy: produce more with less. CIRP J the uncertainty pertaining to available and required pro- Manuf Sci Technol 1:45–52 4. Scholz-Reiter B, Morosini Frazzon E, Makuschewitz T (2010) duction capacities and process times cannot be handled by Integrating manufacturing and logistic systems along global conventional Mathematical Programming and Operations supply chains. CIRP J Manuf Sci Technol 2:216–223 Research approaches. 5. Viswanadham N, Gaonkar RS (2003) Partners selection and Going beyond the configuration and planning phases, synchronized planning in dynamic manufacturing networks. IEEE Transact Robot Autom 19:117–130 further features would include options for lot sizing within 6. Speranza MG, Ukovich W (1994) Minimizing transportation and the DMN as well as options for expanding the use of the inventory costs for several products on a single link. Oper Res proposed approach in the domain of the manufacturing 42:879–894 scheduling, where detailed process plans and configura- 7. Theattre K, Graves S (2001) Tactical shipping and scheduling at Polaroid with dual lead-times. In: Innovation in manufacturing tions have to be considered at each partner’s level. Inte- systems and technology (IMST) report, http://web.mit.edu/ grating data from the shop floor and the logistics network sgraves/www/SMA%20paper%20Oct%2001.pdf. Accessed 3 for monitoring the operation of a DMN is also another idea August 2012 that is worth experimenting with. More sophisticated sce- 8. Seifert RW, Langenberg KU (2011) Managing business dynamics with adaptive supply chain portfolios. Eur J Oper Res 215:551– narios may also be tested, involving the transportation activities and organizations as part of the DMN. 9. Vidal CJ, Goetschalckx M (2001) A global supply chain model DMNs are expected to be in charge of an increasing part with transfer pricing and transportation cost allocation. Eur J of the global manufacturing activity and therefore provid- Oper Res 129:134–158 10. Klincewicz JG, Luss H, Yu CS (1988) A large-scale multiloca- ing new methods and tools for improving their operation, tion capacity planning model. Eur J Oper Res 34:178–190 and overall efficiency is of paramount importance. 11. Stadtler H (2009) A framework for collaborative planning and state-of-the-art. OR Spectrum 31:5–30 Acknowledgments The work in this paper has been partially sup- 12. Bhatnagar R, Chandra P, Goyal SK (2004) Models for multi-plant ported by the FP7 Integrated Project ‘‘IMAGINE—Innovative end-to- coordination. Eur J Oper Res 67:141–160 end Management of Dynamic Manufacturing Networks,’’ funded by 13. Pibernik R, Sucky R (2007) An approach to inter domain master the CEU. planning in supply chains. Int J Prod Econ 108:200–221 14. Dudek G, Stadtler H (2005) Negotiation-based collaborative planning between supply chain partners. Eur J Oper Res 163: 668–687 References 1. Chryssolouris G (2006) Manufacturing systems—theory and practice. Springer, New York

Journal

Logistics Research – Springer Journals

Published: Sep 14, 2012

There are no references for this article.