This volume, edited by Wolfgang Kersten and Thorsten Blecker, offers the most important perspectives on supply chain risk management. The contributions written by named experts provide actual information about workable approaches for supply chain risk management, analyses of supply chain risks, identification of key risk factors for logistics outsourcing, assessment of the uncertainty of delivery. With this book readers will gain central insights how to handle approaches for supply chain risk management within their business. They will learn how to manage risks effectively to build leaner supply chains with a maintainable risk exposure for all partners in industry and services.

Intelligent logistic objects with the ability of autonomous control are a possibility to face new challenges in dynamic logistic systems. On the other hand autonomy of logistic entities poses new questions to the underlying logistic system. In this paper we will focus on the possibilities of reducing the objective and subjective drawbacks of autonomy in logistics by an autonomous risk management for the logistic entity. A technical solution based on intelligent agent technology will be presented.

Autonomous logistic processes aim at coping with logistic dynamics and complexity by local decision making to gain flexibility and robustness. This paper discusses resource-bounded logistics decision making using software agents and task decomposition. Simulations show the feasibility of dynamic vehicle routing and quality monitoring on embedded systems for the transport of perishable goods.

Globally distributed production networks accompanied by the reduction of the vertical range of manufacturing, customer-driven markets, decreasing product life-cycle times and increasing information flows alter the requirements for the management of logistic systems and processes. The reduction of the size of goods that have to be transported and as a consequence thereof an increasing amount of transports are main reasons for a relative shortage of logistic infrastructure and lead also to rising utilization of existing logistic processes and to more complex logistic systems. These developments for example are caused through the evolution of virtual organizations and the increasing maturity of new information and communication technologies (ICT) technologies like RFID and ubiquitous computing.

Decision making in a real-world domain like logistics is challenging for an autonomous technical system like a software agent.In this paper the problem of planning in such an environment is addressed.Classical planning and probabilistic criteria-directed scheduling components are tied together by a metalevel control and supplemented by a sophisticated world model and a risk management module to form a plan-based decision support system for autonomous control of logistic entities.The system is designed to be integrated in a multi-agent based simulation for evaluation and will later be used to support autonoumous decision making in real-world logistic domains.

This paper introduces a decision-making model for autonomous cooperating logistics processes (ALPs) in transport organizations, an emerging technology following the paradigm of self-organisation from an interdisciplinary perspective. INTRODUCTION The desire and possibility to have goods available nearly anywhere at any time have contributed to a tremendous increase in transport volume and in delivery frequencies. Customer expectations, the pressure of competition on turbulent global markets, and virtualisation of logistic companies results in complex, dynamic logistic systems, structures, and networks. A promising emerging approach is the use of the self-organisation paradigm for logistic processes (Scholz-Reiter, Windt & Freitag, 2004) to create autonomous logistic processes (ALPs). This approach operates as a direct challenge to central planning as it decentralises control, planning, information, and decision-making in production and transport logistics through the use of innovative information and communication technologies. The main goal of ALPs is to increase a logistic system’s robustness, flexibility, and reactivity. Logistic systems are regarded as socio-technical systems with different sub-systems and levels. This paper focuses on the social and technical decision-making systems of ALPs from an interdisciplinary perspective involving insights from an emerging Sustainable Management approach in economics, on the one hand, and multiagent-systems (MAS) research (Wooldridge & Jennings, 1995), a predominant paradigm in distributed artificial intelligence, on the other. Our purpose is twofold: First, we analyse the problem of delegating decisions from a social decision-making system to an agent-based, i.e., technical, decision-making system and offer an interdisciplinary model for autonomous cooperating logistics processes. Second, we present a multiagent-system-based approach which integrates knowledge management and risk management. In the last part of our paper, we offer implications for further research. SUSTAINABLE MANAGEMENT AS A DECISION-MAKING FRAMEWORK The concept of sustainability becomes increasingly important for transport logistics organizations (McIntyre, 2003; Wu & Dunn, 1995). The value of a sustainability perspective for transport logistics is to control for unintended side effects, long-term feedback effects, and to ensure an organisation’s long-term access to economic, natural, and social resources (Ehnert, Arndt & Müller-Christ, in press). We understand organisations as resource-dependent socio-economic systems, which consume and supply resources. Sustainability is regarded as a rationale to deal with these resources (Müller-Christ, 2001). In the future, new technologies in ALPs will allow transport organizations to delegate decisionmaking competence partly from the social technical decision-making systems (e.g. managers, schedulers) to an agent-based, i.e., technical decision-making system (see figure 1). While strategic decisions remain on the social systems level of an organization operational decisions and goal achievement can be transferred to the technical-operational level. This delegation from a centralized social and technical management system to autonomous units (agents) is accompanied by various challenges on both the social and technical levels. For example, it is a difficult task to guarantee the sustainability of decisions, if there is no longer a unique central organisational unit where all the necessary information and experience is available. Therefore, we apply a newly developed Sustainable Management framework for ALPs which encompasses a management of organizational dilemmas (dilemma management), of organizational boundaries (boundary management), and a management of participation (participation management) (see Ehnert et al., in press). This framework provides the basis for the delegation of decisions from the social to the technical systems level and it is influenced by feedback effects of delegation as e.g. higher degree of goal achievement (see figure 1). Figure 1: Interdisciplinary model of delegating decision-making for autonomous cooperating logistics processes The objective of a Sustainable Management is to provide logistics managers with a frame to analyse and design the particular management situation in their organization, to profit from the full potential of ALPs, and to support a sustainable development of the organization. Specified solutions are not Knowledge Management for and with technical systems Risk Management for and with technical systems Sustainable Management social decisionmaking system technical decision-making system Dilemma Management Boundary Managemen t Participation Management autonomous identification generation of alternatives online optimisation handle & inference acquisition & update process & learn Feedback effects of delegation on Sustainable Management Framework Delegation of decisionmaking competence provided in this approach due to the contextual differences between organizations. A dilemma management is needed for ALPs as transport logistics organizations have to reconcile the dilemmas of ‘central versus autonomous control’ and of ‘efficiency versus sustainability’ (see Ehnert et al., in press). Dilemma situations require a choice between two equally important and contrary alternative actions (Neuberger, 2002). Reconciliation strategies for ALPs encompass temporal, spatial, and spherical separation and synthesis of dilemmas (Ehnert et al., 2006). In this paper, we suggest to mix different reconciliation strategies. For the dilemma of central versus autonomous control, we suggest to combine spatial and spherical separation because of the nature of ALPs. ALPs separate the dilemma of central versus autonomous control spatially, shifting the decision-making to different locations in and between organizations. The result is a tension between the social and technical decision-making systems. This tension can be used constructively by spherical separation, i.e. by addressing the poles of the dilemma simultaneously in the same subsystem. On the social decision-making level, the dilemma of central versus autonomous control could be addressed by an agent decision mechanism which incorporates both, the ability to act autonomously based on the proposed riskand knowledge management framework, and to interact with a central authority whenever appropriate and possible. The level of autonomy an agent can achieve or may exercise is subject to further research (see Timm, 2006 and conclusions). For organizations striving for sustainability, ALPs offer the advantage of collecting data with the help of an agent-based knowledge and risk management and thus reducing the danger of unintended feedback loops on the transport organization. The purpose of a boundary management is to support the regulation of boundaries within and between organizations that implement ALPs. This is important, because ALPs presuppose new relationships in and between transport logistics organizations requiring them to open their boundaries. These relationships have to be long-lasting if logistics corporations want to achieve their goals efficiently and at the same time secure their long-term success and continued existence. A boundary management has to address relevant organisational decision premises (cp. Luhmann, 2002) in order to manage the process of boundary opening (Ehnert et al., in press). For example, reflecting, negotiating and implementing collective strategies (cp. Astley & Fombrun, 1983) can be an important aspect of managing boundaries between transport logistics organisations which cooperate via ALPs. As boundary openings come along with an increased vulnerability (e.g. towards competitors using shared information for their individual benefit), the linking of cooperation and strategic intent may motivate organisations to refrain from short-sighted opportunistic behaviour and provide the necessary basis for the willingness to open organisational boundaries. In further research, this dilemma and boundary management is going to be complemented by a participation management because ALPs require that all important stakeholders (e.g. employees, transport logistics partners) are actively involved in the process. Maintaining a Sustainable Management in a highly dynamic and distributed environment presupposes complex social mechanisms of coordination in order to enable a flexible and proactive knowledge management, and an elaborated risk management for identifying and assessing risks as precisely as possible in dynamic and complex environments. This objective can be achieved with the help of a multiagent-system-based approach. KNOWLEDGE MANAGEMENT IN A MULTIAGENT-BASED APPROACH In our framework, agents are used to represent real world logistic entities such as trucks and containers, abstract objects such as weather or traffic services, or even human decision makers, such as a ramp agent at a loading dock. Our approach to knowledge management consists of three main components: conceptual knowledge, roles, and parameters (Langer et al., 2005). The conceptual knowledge is represented as an OWL (Web Ontology Language, cf. http://www.w3.org/2004/OWL/) ontology. For the purpose of our logistic application domain, this ontology includes a representation of the transportation or production network, the basic types of agents and their properties (e.g., for a vehicle, its average and maximum speed, the types of routes in the network it can use, and its load capacity), and the properties of 'inactive' objects, such as highways, depots, etc. In contrast to previous approaches to agentbased knowledge management, we do not presuppose a one to one correspondence between agents and knowledge management functions, such as providing knowledge or brokering knowledge. In our approach these functions are implemented as roles. A knowledge management role includes certain reasoning capabil

The concept of autonomous logistic processes addresses the emerging requirements in current and future logistics by applying the latest information and communication technologies. They enable autonomous systems that operate and cooperate as lo cal representatives of logistic entities. The analysis and design of these processes is subject t o simulation studies. Two simulation systems for the analysis of autonomy in logistics ‐ an agent-ba sed and a discrete event approach ‐ are presented. Inspired by the time concept of discrete ev ent simulation, a new synchronisation technique is proposed that allows for a dynamic adaptation of simulation time progression in multiagentbased simulation depending on the granularity curre ntly needed.

The current trends and recent changes in logistics lead to new, complex and partially conflicting requirements on logistic planning and control systems. Currently available strategies and methodologies do not address these new requirements sufficiently. The concept of autonomous logistic processes intends to overcome these drawbacks together with latest information and communication technologies. Their analysis and design is subject to simulation studies. Two simulation systems for the analysis of autonomy in logistics with an agent-based and a discrete event approach are presented. Both systems are designed and suitable for different aspects of autonomous logistic processes.

The trends and recent changes in logistics lead to complex and partially conflicting requirements on logistic planning and control systems. Due to the lack of efficiency of currently available strategies and methodologies, a new paradigm for logistics planning and control is required. An emerging approach is the analysis and design of autonomous logistic processes. Agents represent a modern approach for implementing autonomous systems. The challenge for the design of agent systems is to integrate the complex and dynamic knowledge required for reliable decision-making in logistics. To address this problem, we introduce a framework for distributed knowledge management in competitive environments. Our approach combines a general role model enabling distributed, flexible agent-based knowledge management services and a set of general decision parameters for rational agents.

Sensitive Goods (i.e. fruits, vegetables, paper rolls, cellulose) need a special treatment in a logistic environment. For this reason the logistic processes have to be planned in consideration of special transportation conditions. In this context, the existence of possible hazards and chances has to be regarded. In order to handle the existing risks of logistic systems and especially for sensitive goods we propose a proactive risk management (RM) system to supplement a holistic process management. It supports the design of processes which are robust and insusceptible to existing and occurring anthropogenic and environmental hazards. This paper analyzes essential parts for a convenient risk definition and examines concepts and tools that allow the adequate management of system and process related risk. INTRODUCTION Modifications of product life cycles, company structures and information flows alter the requirements for logistic processes. Logistic processes are facing new challenges. These are results by the development of virtual enterprises and the increasing maturity of new ICTs like RFID and ubiquitous computing. The rising complexity of organizational structures leads to a mounting utilization of existing processes. To coordinate all of these processes an increasing demand of required information for just in time deliverables is needed. These requirements exceed the abilities of existing standard logistic processes. This can be realized by the development of autonomous, decentralized control systems, which select alternatives autonomously and decide within a given framework of goals. Experiences show, that a high number of autonomously acting objects lead to an increased sensitivity and higher risk. Direct disturbances of the processes caused by anthropogenic risks and natural hazards have to be identified and reduced by a pro active RM system. The RM system has to regard the transport conditions and the goals of the logistic processes as well as the requirements for the objects which will be transported. The requirements for the transport of the logistic object depend on character of this logistic object. Some objects need a special treatment during the whole transport because they are sensitive and susceptible to damage. This paper will deal with the particular requirements of sensitive goods in a logistic environment of autonomous logistic objects. The content of this paper and the treated problems are the base for this scenario: A load of fish has to be transported in a truck equipped with a light cooling system. The payload and the truck are represented by agents. While the truck is in a traffic jam on the way from A to B, the agent of the payload perceives that the temperature is getting higher. As a consequence the risk for payload to get spoiled before the truck will reach its goal under this traffic conditions is getting to high. For this reason the agent evaluates a possibility with a lower risk within the framework of goals. He decides that the payload will have to be Bemeleit, B.; Lorenz, M.; Schumacher, J.; Herzog, O.: Risk Management for Transportation of Sensitive Goods. In: Proceedings of the 10th International Symposium on Logistics (10th ISL). 2005, pp. 492-498

We present a novel approach to enable decision-making in a highly distributed multiagent environment where individual agents need to act in an autonomous fashion. Our architecture framework integrates risk management, knowledge management, and agent deliberation to enable sophisticated, autonomous decision-making. Instead of a centralized knowledge repository, our approach supports a highly distributed knowledge base in which each agent manages a fraction of the knowledge needed by the entire system. Our approach also addresses the fact that the desired knowledge is often highly dynamic, context-sensitive, incomplete, or uncertain. Thus risk management becomes an integral component which enables context-based, situation-aware decision making, which in turn supports autonomous, self-managing behavior of the agents. A prototype system demonstrating the feasibility of our approach is being developed as part of an ongoing funded research project.

We present a novel approach to enable decision-making in a highly distributed multiagent environment where individual agents need to act in an autonomous fashion. Our architecture framework integrates risk management, knowledge management, and agent deliberation to enable sophisticated, autonomous decision-making. Instead of a centralized knowledge repository, our approach supports a highly distributed knowledge base in which each agent manages a fraction of the knowledge needed by the entire system.

We present a framework for role-based knowledge manage- ment in a multiagent environment. Our approach focuses on roles which are carried out by rational agents. The use of roles for knowledge man- agement, which is orthogonal to the organizational entity represented by an agent, reduces the computational cost of reasoning and simplifies the agent model. Our approach and illustrative examples are couched in the context of the logistics domain. Continued and strong demand for increased customization of products and their delivery has brought about a sea change in the economical landscape: from mar- kets that were predominantly controlled by sellers to markets that are now driven by buyers and their demands. To meet the resulting requirements in the logistics domain , participating enterprises investigate ways to restructure their business processes to allow for more autonomy in order to provide flexibility and rapid response when reacting to customer requests. Such restructuring away from the traditional centralized way of doing business is made possible by an emergence of hardware technologies including GPS-based telematics for trucks, more reliable and longer ranging wireless communication as well as low-power sensor devices. In addition, innovative software is being developed to support autonomous decision-making and to provide the right information to the right processes when it is needed. Software systems implementing autonomous logistic processes (e.g., agents) need to share information on a continuous basis, for example, prod- uct specifications, manufacturing capabilities, delivery schedules, etc., and are required to make decisions which are consistent with the policies and overall economical situation of the enterprise they represent. In addition, agents must be able to negotiate, form coalitions, and thrive in the presence of competition, and are also subject to unpredictable changes in their environment. Furthermore the dynamics of logistic processes require the ability to plan (re-plan) even in light of uncertainty, incomplete, or false knowledge. Stan- dard scenarios of logistic processes typically have been modeled on the basis of static graph-theoretic representations. The well-known traveling salesman prob- lem (TSP), the vehicle routing problem (VRP), or the pickup & delivery problem

The overall aim of this work is to grant blind users access to graphically represented information. In order to enable them to also search and retrieve this information an RDF(S) representation is shown which further leads to an application which enables another tininess of the semantic web by extracting explicit semantics of line drawing images.