A web-based decision support system for waste lube oils collection and recycling

https://doi.org/10.1016/j.ejor.2007.11.004Get rights and content

Abstract

This paper presents a web-based decision support system (DSS) that enables schedulers to tackle reverse supply chain management problems interactively. The focus is on the efficient and effective management of waste lube oils collection and recycling operations. The emphasis is given on the systemic dimensions and modular architecture of the proposed DSS. The latter incorporates intra- and inter-city vehicle routing with real-life operational constraints using shortest path and sophisticated hybrid metaheuristic algorithms. It is also integrated with an Enterprise Resource Planning system allowing the utilization of particular functional modules and the combination with other peripheral planning tools. Furthermore, the proposed DSS provides a framework for on-line monitoring and reporting to all stages of the waste collection processes. The system is developed using a web architecture that enables sharing of information and algorithms among multiple sites, along with wireless telecommunication facilities. The application to an industrial environment showed improved productivity and competitiveness, indicating its applicability on realistic reverse logistical planning problems.

Introduction

This paper reports on a research project with a leading multi-national Greek company in lubricant production and sole in recycling in Greece. The motivation for this research stems from the interrelationship between environmental considerations and benefits of recycling products. We consider waste lube oils (WLO) collection and regeneration as the focal coupling process of environmental consciousness and profitable recycling operations, and propose the appropriate decision making methodologies, management tools and techniques to transform these processes into state-of-the-art seamlessly running systems. The main focus is on the development of a web-based decision support system (DSS) that enables schedulers and decision makers to address reverse logistical planning problems and monitor effectively operations using novel computational methods and Information and Communication Technology (ICT).

The WLO collection and regeneration processes involve the upstream movement of WLO from several sets of collection points through a network of accumulation points to a central regeneration facility, where new lubricants are produced. Accumulation points refer to regionalized intermediate transfer facilities used to accumulate WLO locally. The latter is the major raw material for the production of basic lube oils and new lubricant products, via a continuous multi-stage regeneration. To the other end, the collection needs for WLO (actual production–regeneration requirements) is determined by customers’ demand for end-item lubricant products. Therefore, distribution of end-items, production scheduling, inventory and WLO collection are all linked in a reverse planning fashion constituting a closed-loop supply chain.

Reverse logistics encompass the activities all the way from used products no longer required by the user to products again usable in market (Fleischmann et al., 1997). Therefore, while logistic activities involve material flow forward through the supply chain, reverse logistics, or similarly closed-loop supply chains, address the return flows from the user and encompass the re-processing into a usable product. There are four basic characteristics, namely the return reasons and driving forces, the type of products, the recovery processes and the actors involved. All these characteristics are interrelated and their combination determines to a large extent, the type of issues arising from the resulting reverse logistics system.

Carter and Ellram, 1998, Fleischmann et al., 1997, Dekker et al., 2003 provide recent surveys on theory building, quantitative modeling and case studies. Beullens (2004) provide an exploratory analysis on several operational and strategic issues penetrating reverse logistics systems. French and LaForge (2006) investigate several re-use issues (returns, re-use options and others) and practices related to process industry firms. Finally, Hu et al. (2002) examine hazardous-waste reverse logistics systems. Although there is a vast literature on reverse logistics, most quantitative models focus on network design (i.e. location of joint facilities, allocation of returning materials to open facilities. etc.) and inventory control. To this end, few consider vehicle routing and scheduling with respect to the reverse collection or distribution activities. Among others, Dethloff (2001) addresses a reverse distribution and collection problem where re-usable packaging and recyclable goods had to be transported in the reverse direction for remanufacturing. Alshamrani et al. (2007) deal with a reverse blood distribution problem of the American Red Cross. Finally, Teixeira et al. (2004) present a case study for the recycling of solid waste in urban areas.

In order to determine the planning requirements of WLO collection and recycling various aspects must be considered. Although, the locations and capacity of the accumulation points network along with the available resources of both forward and reverse channels are pre-specified, there is an interaction among them affecting both short and long term planning. More precisely, the customers’ demand for lubricant products partly determine the Master Production Schedule (MPS), and consequently, the Material Requirements Planning (MRP). To this end, the MRP along with the sales forecasts and the inventory policy determines the WLO daily collection requirements. Although inventory plays the buffer role between production and collection, practically, one has to devise a production schedule that allows for a compromise between production effectiveness, inventory handling costs and collection vehicle resources. On the other hand, production planning partly determines distribution schedules of lubricant products to final customers.

From the operational viewpoint, on a daily basis, given the desired production plans and the current inventory status, the scheduler has to determine the collection requirements and define which collection and accumulation points will be serviced. Subsequently, the scheduler has to determine the associated collection routing plans. In particular, sets of collection vehicle fleets located at accumulation points, collect (pickup) WLO from collection and accumulation points and transport them either to the respective accumulation points or to the central storage facility of the regeneration plant. Similarly, the scheduler has to determine the daily distribution plans of lubricant products to customers. To-date, planning is empirically performed, without considering the MPS, inventory, availability of resources and other important factors. Thus, a web-based DSS is developed to address planning and monitoring needs of both collection and distribution operations. The latter need is further reinforced by the explosive number of the collection points participating in the recycling operations and the strict laws imposing the collection of used lubricants.

The rapidly growing interest on developing DSS and the appropriate OR methodologies and tools that could aid logistics managers to lower costs and achieve greater flexibility, has attracted significant attention from researchers (Tarantilis et al., 2005). The advances in ICT and the emergence of accurate spatial information systems have provided all technical prerequisites that enable companies to integrate vehicle fleet routing and scheduling with other key functions. In the context of modern telecommunication systems, Electronic Data Interchange (EDI), Global Positioning Systems (GPS) and Geographic Information Systems (GIS) are the main enablers of fleet management within enterprise wide end-to-end information systems.

Generation of vehicle routing plans require access to spatial sub-systems, network optimization procedures and particular geo-processing spatial tools provided by GISs using specialized spatial data storage formats. The combination of DSS and GIS functions provides the framework for vehicle routing DSS as they synthesize decision making tools and routing algorithms with spatial tools including the ability to provide visual feedback in the form of maps. Such systems are also referred as Spatial decision support systems (SDSS)(Sharma et al., 2006). A wide variety of vehicle routing DSS appears in literature, including distribution systems (Tarantilis and Kiranoudis, 2001, Tarantilis and Kiranoudis, 2002a, Tarantilis and Kiranoudis, 2002b), collection (Butler et al., 2005), dispatch and delivery (Weigel and Cao, 1999) and ship scheduling (Fagerholt, 2004, Sambrakos et al., 2004). Other more recent examples of vehicle routing DSS can be found in Belenguer et al., 2005, Tarantilis et al., 2004b, Ioannou et al., 2005, Ruiz et al., 2004.

Most of the aforementioned DSSs fall into the category of model-driven single-user systems, where core routing problems are modeled and solved using optimization methods supported by peripheral spatial tools (Ray, 2005). To the other end, given the increasing need to better support group decision making, research moves towards to larger multi-user enterprise wide systems which facilitate centralized–regionalized vehicle routing DSSs. As stated by Hall (2006), centralization is allowing companies to aggregate data and monitor the overall performance. Obviously, such capabilities and integration with other peripheral functional modules are mainly being provided through web-based systems that enable sharing of information and algorithms among multiple sites. Additional advantages of web-based DSS include platform independence, shorter learning curves for users already familiar with web tools, lower software distribution costs and ease of performing system updates (Tarantilis et al., 2008).

The movement towards web-based DSSs allows planning or optimization tools to be distributed more widely, since all decision support related operations are performed on network servers. Recently, Prindezis and Kiranoudis (2005) developed an internet based logistics management system to co-ordinate and disseminate tasks and related information for a real life meat distribution system. Ray (2005) developed a web-based DSS for managing the movement of oversize-overweight vehicles over highways. Finally, Tarantilis (2006) proposes a vehicle routing DSS developed in a way that allows it to be used as a web-based Application Service Provider (ASP). Similar to internet service providers, that linked businesses and consumers up to the internet, ASPs lease software applications to businesses and consumers via the internet.

In this paper, the systemic dimension and the modular architecture of the proposed web-based DSS is illustrated. Emphasis is given on the integration of online monitoring and reporting with other planning tools such as MRP, inventory control, shortest path, vehicle routing and their respective interactions and interrelationships. Particularly, the way to setup efficient collection and distribution channels is analyzed, along with the decisions that have to be made with respect to the actors participating (collection & accumulation points) and the possible interaction between production and collection processes. Additionally, for both WLO collection and lubricant distribution problems, sophisticated hybrid metaheuristic methodologies are proposed and incorporated to the DSS. Finally, the system design and implementation experience to an actual industrial environment is reported, while configuration, testing and system benefits are also discussed.

The remainder of the paper is organized as follows: Section 2 presents all aspects and operational realities of WLO collection, recycling and lubricant distribution operations. Section 3, illustrates the proposed web-based DSS system structure followed by a detailed presentation of all functional modules. The emphasis is on the systemic dimensions, while a thorough analysis of the mechanisms and attributes of the integrated OR methodologies and tools is provided. Section 4, discusses the characteristics of the case study company, the project phases and the progress of implementation along with the immediate and future benefits. Finally, the paper concludes in Section 5, offering also pointers for further research.

Section snippets

Upstream movement of WLO – collection stages

Waste lube oils can be found at several places, such as gas stations, automobile workshops, production sites, industrial facilities and others, all referred as collection points. The WLO collection and regeneration processes involve the upstream movement of WLO from several sets of collection points to intermediate transfer accumulation points and to the central regeneration and storage facility (central accumulation point), where new lubricants are produced. Fig. 1 illustrates the movement of

Decision support system

WLO collection and regeneration are complex processes, and the proposed DSS has to address all issues pertaining the production, inventory, collection and distribution functions. The four major questions that the DSS must tackle are: (i) how much WLO must be available at the central accumulation point in order to allow for smooth production; (ii) how to effectively schedule the collection process performed by heterogeneous vehicle fleets, considering all operational constraints; (iii) how to

Motivation

According to the European Union (EU) directive 75/439/EEC1, all countries members are obliged, whenever possible, to collect and regenerate WLO in an effort to ensure environmental protection and public health. Despite the environmental consequences, historical data indicates that WLO collection and regeneration in Greece is performed at the lowest rate within EU-15 (see Table 1). However, it is worth mentioning that Greece has one

Conclusions

This paper presented the systemic dimension and the modular architecture of a web-based DSS that uses sophisticated optimization methods to manage effectively reverse logistical planning problems. The DSS is developed using web technology that allowed sharing of information and algorithms. The focus was on developing the appropriate planning tools that enable a WLO regeneration company to improve substantially and monitor effectively the overall performance of its production and collection. The

Acknowledgements

The author is indebted to the anonymous reviewers for their useful comments and suggestions that helped improve the work presented in this paper.

References (46)

  • R. Ruiz et al.

    A decision support system for a real vehicle routing problem

    European Journal of Operational Research

    (2004)
  • C.D. Tarantilis et al.

    A meta-heuristic algorithm for the efficient distribution of fresh meat

    Journal of Food Engineering

    (2001)
  • C.D. Tarantilis et al.

    Using a spatial decision support system for solving the vehicle routing problem

    Information & Management

    (2002)
  • C.D. Tarantilis et al.

    Distribution of fresh meat

    Journal of Food Engineering

    (2002)
  • C.D. Tarantilis et al.

    A threshold accepting meta-heuristic for the heterogeneous fixed fleet vehicle routing problem

    European Journal of Operational Research

    (2004)
  • C.D. Tarantilis et al.

    Combination of geographical information system and effective routing algorithms for real life distribution operations

    European Journal of Operational Research

    (2004)
  • C.D. Tarantilis et al.

    A webbased ERP system for business services and supply chain management

    Application to real-world process scheduling. European Journal of Operational Research

    (2008)
  • C.D. Tarantilis et al.

    A flexible adaptive memory based algorithm for real-life transportation operations: Two case studies from dairy and construction sector

    European Journal of Operational Research

    (2007)
  • J. Teixeira et al.

    Recyclable waste collection planning-a case study

    European Journal of Operational Research

    (2004)
  • Y. Bassok et al.

    A simple linear heuristic for the service constrained random yield problem

    IIE Transactions

    (2002)
  • P. Beullens

    Reverse logistics in effective recovery of products from waste materials

    Reviews in Environmental Science & Biotechnology

    (2004)
  • O. Bräysy

    A reactive variable neighborhood search for the vehicle routing problem with time windows

    INFORMS Journal on Computing

    (2003)
  • C.R. Carter et al.

    Reverse logistics: A review of the literature and framework for future investigation

    International Journal of Business Logistics

    (1998)
  • Cited by (56)

    • A variable neighborhood search based matheuristic for a waste cooking oil collection network design problem

      2022, European Journal of Operational Research
      Citation Excerpt :

      The author first estimates the amount of waste oil to be collected and then determines the number of routes per year to visit the communities and the time required to travel to the waste oil producers to fill and empty the truck. Repoussis, Paraskevopoulos, Zobolas, Tarantilis, & Ioannou (2009) develop a web-based decision support system for a waste lube oil collection and recycling network. Vehicles are assumed to be compartmentalized in order to avoid different quality oils getting mixed.

    • An adaptive large neighborhood search algorithm for a selective and periodic inventory routing problem

      2014, European Journal of Operational Research
      Citation Excerpt :

      Finally, Section 6 provides concluding remarks and discusses possible directions for future work. The economic and environmental impact of recycling and remanufacturing of end-of-life products has led to a number of studies in the OR literature (see e.g. Van Wassenhove, Beullens, & Van Oudheusde (2004), Bloemhof-Ruwaard, Fleischmann, Beullens, & Van Wassenhove (2001), Antunes, de Souza, & Teixeira (2004), Zobolas, Tarantilis, Ioannou, Repoussis, & Paraskevopoulos (2009), Benjamin & Beasley (2010)). Belien, De Boeck, and Van Ackere (2011) reviewed solid waste collection problem types with a particular focus on vehicle routing problems in the literature, and give a classification of the solution methods proposed for each problem type.

    View all citing articles on Scopus

    This work is supported by the General Secretariat for Research and Technology of the Hellenic Ministry of Development under contract GSRT NM-67.

    View full text