Fink, Andreas

Battery electric freight vehicles have the potential to mitigate the local urban road freight transport emissions, but their numbers are still insignificant. Logistics companies often consider electric vehicles as too costly compared to vehicles powered by combustion engines. The current literature suggests that increasing the mileage can maximize the competitiveness of electric freight vehicles. In this manuscript we develop a generic model to determine the cost-optimal balance between a high utilization of electric freight vehicles – which often have low operational costs – and their required expensive battery replacements. Our work relies on empirical findings of the real-world energy consumption from a large German field test with vehicles of 7.5 and 12 tons, respectively. Our results suggest that increasing the range to the technical maximum by intermediate (quick) charging and multi-shift usage is not the most cost-efficient strategy in every case. A low daily mileage is more cost-efficient at high energy prices or consumptions, relative to the diesel prices or consumptions, or if the battery is not safeguarded by a long battery warranty. In practical applications our model may help companies to choose the most suitable electric vehicle for the application purpose, or the optimal trip length from a given set of options. For policymakers, our analysis provides insights on the relevant parameters that may either reduce the cost-gap at lower daily mileages, or increase the utilization of electric urban freight vehicles, in order to abate the negative impact of urban road freight transport on the environment.

Decision making in operational planning is increasingly affected by conflicting interests of different stakeholders such as subcontractors, customers, or strategic partners. Addressing this, automated negotiation is a well-suited mechanism to mediate between stakeholders and search for jointly beneficial agreements. However, the outcome of a negotiation is strongly dependent on the applied negotiation protocol defining the rules of encounter. Although protocol design is well discussed in literature, the question on which protocol should be selected for a given scenario is little regarded so far. Since negotiation problems and protocols are very diverse, the protocol choice itself is a challenging task. In this study, we propose a decision support system for negotiation protocol selection (DSS-NPS) that is based on a machine learning approach – an artificial neural network (ANN). Besides presenting and discussing the system, we, furthermore, evaluate the design artifact in elaborate computational experiments that take place in an intercompany machine scheduling environment. Our findings indicate that the proposed decision support system is able to improve the outcome of negotiations by finding adequate protocols dynamically on the basis of the underlying negotiation problem characteristics.

The increasing coupling of planning and scheduling between different companies leads to novel challenges in devising and implementing effective decision support systems. In this paper, we describe a hard decentralized scheduling problem with heterogeneous machines and competing job sets that belong to different self-interested stakeholders (agents). These agents want to minimize their costs that consist of individual tardiness cost as well as their share of the machine operating cost. The determination of a beneficial solution, i.e., a respective contract in terms of a common schedule, is particularly difficult due to information asymmetry and self-interested behavior of the involved agents. To solve this coordination problem, we present two automated negotiation protocols with a set of optional building blocks. In the first protocol, new solutions are iteratively generated as mutations of a single provisional contract and proposed to the agents, while feasible rules with quotas restrict the acceptance decisions of the agents and, thus, the successive adaptation of the provisional contract. The second protocol is based on a population of contracts and mimics evolutionary processes. For evaluation purposes, we built a simulation testbed and conducted computational experiments. The computational study shows that the protocols can achieve high quality solutions very close to results from centralized multi-criteria procedures. Particular building block configurations yield improved outcomes, e.g., in case that the agents are also allowed to make contract proposals. Thus, the presented approach contributes to the methodology and practice of collaborative decision making.