Integration of logistics outsourcing decisions in a green supply chain design: A stochastic multi-objective multi-period multi-product programming model

Highlights

• Integration of some critical Supply Chain Management issues in the green supply chain design.

• Suggestion of constructive models to roughly estimate logistics costs and carbon emissions.

• An example of stochastic plan is provided to capture different business uncertainties.

• An Epsilon-constraint algorithm leads to a set of Pareto optimal green configurations, with optimal levels of logistics outsourcing.

Abstract

This paper develops a programming model, which combines logistics outsourcing decisions with some strategic Supply Chains’ planning issues, such as the Security of supplies, the customer Segmentation, and the Extended Producer Responsibility. The purpose is to minimize both the expected logistics cost and the Green House Gas (GHG) emissions of the Supply Chain (SC) network, in the context of business environment uncertainty. First, we define a general structure of the closed-loop SC. Second, we provide constructive models to roughly estimate the insourcing and outsourcing logistics costs, and their corresponding GHG emissions. Third, we establish a stochastic plan based on a scenarios approach to capture the uncertainty od demand, capacity of facilities, quantity and quality of returns of used products, and the transportation, warehousing, and reprocessing costs. Fourth, we suggest a programming model, and an algorithm based on the Epsilon-constraint method to solve it. The result is a set of optimal non-dominant green SC configurations, which provide the decision’ makers with optimal levels of logistics outsourcing integration within a decarbonized Supply Chain, before any further low-carbon investment.

Introduction

Supply Chains (SCs) involve Suppliers, Manufacturers, Distributors & Retailers, Consumers, and other partners such as Third-Party Logistics providers (3PLs) and Recyclers. Each link in a SC while adding value to the products, contributes to degradation of the natural environment; particularly by the climate change problem involvement (Dasaklis and Pappis, 2013). Therefore, the decisions regarding the activities performed by the mentioned actors will determine both the environmental and the economic performances of the SCs (Wang et al., 2011):

Concerning the environmental performance, Huang et al. (2009) have reported that more than 75% of the Green House Gases (GHG) emissions of many industry sectors come from their SCs. So, reducing those indirect GHG emissions may be more cost-effective for an industrial company, than reducing its direct GHG emissions (Montoya-Torres et al., 2015). In Browne et al. (2009), the World Economic Forum suggests thirteen effective strategies to decarbonize the SCs, and among the most effective ones: Improving the network logistics planning, through global optimization.

Concerning the economic performance, and according to 19th, and 17th annual 3PL studies of Langley & Capgemini (2015; 2013) the total of logistics expenditure of the eight largest industry sectors in the world is between 12% and 15% of the sale revenue, and about 40% of the global logistics activities is outsourced to the 3PLs. The most important logistics activities outsourced are freight transportation, warehousing, and reverse logistics. According to the authors, the logistics outsourcing, as a flexible strategy can reduce logistics costs by 10%, logistics fixed-asset by 15%, and inventory by 25%, if it is well defined by the focal company (FC). So, considering the possibility of 3PL integration within the SC is of great importance to minimize the costs, and reduce the business risks (Jayaram and Tan, 2010).

However, the 3PLs seem not undertake concrete sustainable initiatives vis-à-vis the energy efficiency, the GHG emissions, and the traffic congestion (Evangelistia et al. 2011; Blanco and Craig, 2009). For instance, the 3PLs tend to use a flexible routing network strategy, rather than a point-to-point strategy, to consolidate the freight of different customers (Hesse and Rodrigue, 2004). This can generate a lot of stops between different origins and destinations, hardly provoke traffic congestion, increase relatively the distances, and therefore raise the GHG emissions.

So, considering the potential economic efficiency of 3PLs, and their presumed environmental inefficiency, two main questions are raised in this paper:

• Given that the freight transportation, warehousing, and reprocessing of reused product for the purpose of remanufacturing are not the FC's core activities, one of the most important decisions to be taken is whether or not outsourcing totally or partially such logistics activities to 3PLs, in the context of a green SC.

• How does the optimality of GHG emissions of logistics activities, and corresponding logistics costs affect the configurations of a closed-loop SC network integrating 3PLs, in the context of business uncertainty?

The main contribution of the present paper is the suggestion of a more realistic programming model, which integrates logistics outsourcing decision within the closed-loop SC design network problem, in the context of business uncertainty. The model captures three important issues of the SC management: (1) The security of supplies, by considering the portfolio model of supplies (Kraljic, 1983); (2) the segmentation of market, for meeting the heterogeneous requirements of customers (Lee, 2002); and (3) the Extended Producer Responsibility for managing effectively the End of Life phase of products (Lindhqvist, 2000).

The objective is to minimize both the expected total logistics cost (e.g. Freight transportation; Warehousing; and Processing returns of used products), and the corresponding expected total GHG emissions, under the constraints of: (a) flow conservation, (b) fleet & facilities capacities, (c) opening of facilities, and (d) installing hybrid facilities, which may be leased and operated by FC, or owned and operated by 3PLs.

• We provide three constructive models to make rough estimate of logistics costs and GHG emissions of logistics operations to be insourced or outsourced;

• We suggest a stochastic plan, based on a scenarios approach (Pishvaee et al., 2008) to capture the uncertainty of demand, quantity and quality of returned products, and the variable costs of logistics operations; and

• We suggest an algorithm based on Epsilon- Constraint method (Mavrotas, 2009), to solve the stochastic bi-objective, multi product, multi-period, and multi-echelon programming problem;

The solutions represent a set of non-dominant green SC configurations; which distinguish the logistics activities that should be performed in-house from those that should be outsourced.

The remainder of this paper is organized as follows. In Section 2, we provide a literature overview on the 3PL’ integration within SCs; and on the Green SC network design problem. In Section 3, we define the general structure of a closed-loop SC, and provide three constructive models to roughly estimate the fixed and variable costs, and the fixed and variable GHG emissions of different logistics operations. In Section 4, we present the modelling and solving approaches of the closed-loop SC design problem. Then, we discuss some managerial insights, which can be deducted from the implementation of an example of the model. Finally, in the Section 5 we draw the conclusion.