An optimization model for automated selection of economic and ecologic delivery profiles in area forwarding based inbound logistics networks

Abstract

Area Forwarding Based Inbound Logistics Networks are used by several large companies whose suppliers are spread widely to reduce costs by consolidating transported goods in an early stage of the transport. Managing material flows in those networks is a complex task, especially if the synergy effects in the main leg shall be used to reduce costs and environmental pollution. One technique to decrease steering overhead is the use of delivery profiles, which provide a fixed delivery frequency for each supplier and ease planning for supply chain partners. The selection of a delivery profile has effects on both economic and ecologic outcome of the transportation system and thus should be done carefully. In this work we present a new mixed integer programming model which is able to simultaneously deal with the complex tariff systems used in area forwarding and delivery profile selection in acceptable computing time. Our solution methodology exploits problem specific structure to decompose the model into several parts. On the basis of an industrial case study we evaluate the planning solutions obtained by our model compared to the plans currently implemented in practise. Results are analysed both in terms of monetary savings as well as optimisation runtime.

Critical assessment on business strategy from aviation to retail industry during COVID-19 Pandemic

“…In such cases companies like Walmart use delivery profiles, which reduces overhead costs. The delivery profile is a fixed delivery frequency for each supplier and it helps supply chain partners to plan easily (Schöneberg et al, 2010). These delivery profiles increase vehicle usage and reduce transportation cost along with less Carbon Dioxide emissions.…”

Validating Time Efficiency of AOSR 2.0: A Novel WMS Planner Algorithm for SMEs, under Industry 4.0

“…In literature, several other researchers have computed the standard computation times on Pentium series computers for similar heuristics-based scenarios but in a different context of location-allocation for vehicle-routing problems, and the average CPU time lies between 0.52s to 8.57s [20], [21] or 0.62s to 10.21s [22] with different approaches. Research suggests that, on average, less than or equal to one second (=< 1s) is considered a standard CPU processing time for inbound logistics [23], regardless of hardware configuration.…”

An Integrated Approach to Tactical Transportation Planning in Logistics Networks

“…In a more tactical context, Schöneberg, Koberstein, and Suhl (2010) propose a similar resource-based model for optimizing the choice of delivery profiles in area forwarding-based networks. In such networks, suppliers are grouped into areas and each area is equipped with a consolidation center run by a logistics carrier.…”

“…The model introduced in this paper, as well as the models by Kempkes, Koberstein, and Suhl (2010) and Schöneberg, Koberstein, and Suhl (2010), are based on capacitated network design formulations (see §1.3.2). An alternative approach to modeling nonlinear transportation tariffs are concave-cost network flows; see Guisewite and Pardalos (1990) for a survey.…”

“…Thus, it does not consider globally interdependent resources, making it possible to encapsulate tariff selection in a local subproblem and allowing for larger transportation networks to be solved. It also differs from the model by Schöneberg, Koberstein, and Suhl (2010), which employs delivery profiles to model replenishment cycles, whereas the present paper is concerned with dynamic planning and also allows for more general networks with multiple levels of intermediate hubs instead of two-layered area forwarding-based networks.…”