عملکرد دینامیکی یک سیستم موجودی ترکیبی با خط مشی کانبان در فرآیند بازسازی

In this paper we study a hybrid system with both manufacturing and remanufacturing. The inventory control strategy we use in the manufacturing loop is an automatic pipeline, inventory and order based production control system (APIOBPCS). In the remanufacturing loop we employ a Kanban policy to represent a typical pull system. The methodology adopted uses control theory and simulation. The aim of the research is to analyse the dynamic (as distinct from the static) performance of the specified hybrid system. Dynamics have implications on total costs in terms of inventory holding, capacity utilisation and customer service failures. We analyse the parameter settings to find preferred “nominal”, “fast” and “slow” values in terms of system dynamics performance criteria such as rise time, settling time and overshoot. Based on these parameter settings, we investigate the robustness of the system to changes in return yield and the manufacturing/remanufacturing lead time. Our results clearly show that the system is robust with respect to the system dynamics performance and the remanufacturing process can help to improve system dynamics performance. Thus, the perceived benefits of remanufacturing of products, both environmentally and economically, as quoted in the literature are found not to be detrimental to system dynamics performance when a Kanban policy is used to control the remanufacturing process.

With the growing concern for environmental protection from the customers’ perspective and stricter environment legislations issued by the governments, more and more world leading companies have begun to pay attention to the strategy of product take back. From a production viewpoint, this strategy covers aspects of environmental production, such as • green manufacturing, • use of natural resources, • recycling, • material re-use, • remanufacturing. In a number of recent papers, many issues regarding reverse logistics have been addressed, such as how to design a product so that it is easy to be disassembled and reused [1], how to make reselling, recovery and/or disposal decisions on product recovery [2]. Moreover, managing such a reverse supply chain involves coping with uncertainties, especially those concerned with the quantity, quality and timing of the returned products [3]. In this paper, we focus on the dynamic analysis of a hybrid inventory system with joint remanufacturing of old products and manufacturing of new products. “Old products” are any products, either end-of-life or end-of-lease, that have been returned from the market place and require remanufacturing to be remanufactured to be “as good as new”. As shown in Fig. 1, the products come back to the system after having been used by the customer. Parts of returned products then enter the remanufacture process whilst the rest go to land-fill. Customer demand is satisfied from the service stock, which is replenished from manufacturing and remanufacturing processes. Since the remanufactured products are “as good as new”, we cannot distinguish them from the manufactured ones as soon as they have been pooled into the serviceable stock site. We will investigate the dynamic performance consequences of integrating the remanufacturing process with the traditional pipeline concerned with the production of new products. The traditional pipeline is modelled using the automatic pipeline, inventory and order based production control system (APIOBPCS), representative of industry practice [4], [5] and [6] and the dynamic characteristics of which are well known [5] and [7]. Thus, the APIOBPCS is a well established benchmark by which to judge the impact of new information and material flow structures. We consider the scenario where a Kanban policy is employed in the remanufacturing process. In our Kanban policy, the remanufacturing process is triggered by a remanufacturing inventory level, which releases a Kanban batch, or container quantity. We have observed such a hybrid inventory system in practice, in a Swedish manufacturing company producing aluminium profiles. The basic material for profiles is alloyed aluminium billets. These are warmed up in an induction furnace to a temperature of 450–500 °C and then forced through a die and the finished profile runs out. In this extruding process, depending on the shape of the die, the aluminium scrap is highly variable but can be up to 30%. It is still a valuable source of re-usable raw material. The manager thus needs to decide how to handle the scrap. It could be collected at a recoverable stock point and afterwards it is re-melted, and subsequently remanufactured, and enters the serviceable stock, along with the newly manufactured billets to form the serviceable billet inventory (cf. Fig. 1). In this case, a PUSH/PULL policy is needed to determine the relative timing and quantity of the manufacturing and remanufacturing orders. Alternatively, the aluminium scrap is collected immediately after the extruding process and after a certain (small) target amount is reached, it enters again into the furnace with the new billets for the extruding process. In this latter case, the management of the return scrap process is analogous to a Kanban policy. Another case exists in the (re)manufacturer of photocopiers [8] where used photocopiers first enter the disassembly process to conduct a number of operations including inspection, cleaning and disassembly itself. Disassembled products go through a remanufacturing process such as repair, upgrading, and testing operations. Finally, in the assembly line, new and remanufactured modules are assembled to form serviceable inventory. Even though both PUSH and PULL strategies can be adopted in this situation, the PULL policy was found to be more cost effective. We highlight how adding a Kanban policy in the remanufacturing process affects traditional pipelines, i.e. orders placed and serviceable inventory level. In particular we consider the impact of the remanufacturing process on APIOBPCS parameter selection and test the robustness of the new hybrid system to uncertainties in return yield rate and remanufacturing/manufacturing lead-times. Our approach is in contrast to much of recent operation research work on remanufacturing and reverse logistics as we aim to study the dynamic behaviour of the hybrid system. As Towill [9] has highlighted, an efficient production control system can only be designed and operated if the dynamic behaviour of the constituent parts is properly understood. Table 1 summarises our control theory approach which we adopt, as have John et al. [7], Evans et al. [5] in analysing a traditional APIOBPCS. In this paper we focus on the time domain and the impact of shock stimuli. In particular we develop transfer functions in the “s” domain and use the classic step input as it helps to develop a “rich picture” of dynamic behaviour [11]. As remanufacturing systems are time-varying, the motivation for this research is to study their dynamic response via an appropriate methodology. There is merit in the study being based around a model which for conventional supply chains has generated considerable insight across the whole spectrum from individual value streams (IBM) to complete market sectors (USA Machine Tools). Furthermore analysis drives synthesis, hence the use of a model is capable of analysis via transfer function techniques. The outcome of such analysis is to narrow down the field to be subsequently studied via simulation and hence generate added insight into system behaviour. Obviously the use of such models for reverse logistics requires making speculative judgements on likely modus operandi, but these are based on proven sound logistics management procedures. The outcome of this research is likely to provide generic concepts which bring together supply chain practice with the special political, cultural, and transport problems associated with reverse logistics processes. In the context of other research undertaken in reverse logistics, Dekker et al. [12] have classified quantitative research into three categories of modelling: 1. distribution, 2. inventory and production, 3. supply chain scope. Our contribution in this paper is in the second category by developing a systems dynamics model of a specific case in reverse logistics which is a hybrid manufacturing/remanufacturing system. Also, while the models highlighted by Dekker et al. [12] are OR oriented, as in Table 1, our approach adopts the control theoretic method. The rest of the paper is organized as follows. Section 2 reviews the literature directly relevant to our paper, where we highlight the lack of dynamic methodologies to model remanufacturing systems. In Section 3 we give a formal definition of our model and derive the corresponding continuous time, Laplace transfer functions of the hybrid supply chain. In Section 4 the dynamic behaviour of the system is examined to trade-off a good parameter setting for the system. We then further investigate the robustness of the system by varying physical parameters, such as lead-times and return yields, which are assumed as fixed in the preceding analysis. Section 5 concludes our findings and provides future study.

In this paper we study a hybrid system with manufacturing and remanufacturing. The inventory control strategy we use in the manufacturing loop is an APIOBPCS policy. In the remanufacturing loop we employ a Kanban policy to represent a typical pull system. We analyse the parameter settings to find nominal, fast and slow values in terms of system dynamics performance criteria such as rise time, settling time and overshoot. We particularly note that the settings for a hybrid system are similar as for a traditional APIOBPCS, that is, Ta=2Tm, Ti=Tw=Tm. Based on these parameter settings, we investigate the robustness of the system to change in return yield and the manufacturing/remanufacturing lead time. The results clearly show that the system is robust with respect to the system dynamics performance and the remanufacturing process can help to improve system dynamics performance. Thus, the benefits of remanufacturing of products, both environmentally and economically, are not to the detriment of dynamic performance. Regarding future research, there are at least two directions worth pursuing. One is to investigate the system dynamics performance when the input is random which is closer to the real-world. Another is the system response in the frequency domain. Because no matter what kind of input signal there is it can always be expressed as the sum of a series of sine waves having different frequencies ω and amplitudes a. On the other hand, a sine wave signal in itself can reflect a variety of trends from economic cycles through to seasonal patterns and finally noise. So frequency response analyses could reveal richer insights from a managerial perspective.