How to frame the sustainable Supply chain network?

Introduction

The primary goal of most businesses or organizations has unarguably been profit maximization (Siekelova et al, 2019). This has prioritized economic variables of several companies to the foreground relegating environmental and other social issues to the background. However, the trajectory upon which businesses are run in recent years seems to be changing towards an embracement of the overall pillars of development which are social, economic and environmental (Chiarini, 2014). Following the increasing concerns from global and national policies like the sustainable development goals (SDG), which specifically encourages the sustainable way of entire human operations, companies have been guided and mandated to operate in more sustainable ways. That is, a way the meets the needs of the current population without compromising that if the future population.

This has been evident in the way raw materials are supplied into companies and finished products are supplied out of companies. In both academia and practice, this new evolution of supply chain has been termed as sustainable supply chain. As a process, this has always been accompanied by sustainable supply chain design. The former is defined in various ways following its prominence in recent business literature. The most popular definition however is from Zhu et al (1980) which states that it is  “the management of material, information and capital flows as well as cooperation among companies along the supply chain while taking goals from all three dimensions of sustainable development, i.e., economic, environmental and social, into account which are derived from customer and stakeholder requirements.”

However, in a more recent publication, the UN Global Compact (2021), supply-chain sustainability is the impact a company’s supply chain can make in promoting human rights, fair labor practices, environmental progress and anti-corruption policies. Sustainable supply chain design then explains the models, approaches and techniques that are employed to predict the most efficient way of sustaining the supply chain. As mentioned Krieger and Sackmann (2018), these models establish the long term operational framework of a firm, hence being a significant contributor to how sustainable the operations of a firm is. It is therefore against this backdrop that this study makes efforts to review literature and analyses a case study on how the two modern design approaches (optimization and simulation) have been applied in this regard.

Literature Review on the use Optimization and Simulation to Design Sustainable Supply Chain

Considering the relevance of optimization and simulation models as a way to design sustainable supply chain in contemporary times, this section basically reviews literature covering use of these two approaches to develop a sustainable supply chain design. In two subsections, the first focuses on optimization whilst the latter covers literature on simulation.

Optimization and Sustainable Supply Chain Design (SSCD)

Optimization has been regarded the determination of the "best available" values of an objective function given a specific domain (or input), which may encompass a variety of various sorts of objective functions and domains (Martins et al, 2021). In effect, its application as a solution model to sustainable supply chain design is clustered. As argued by Schreiber (2019), optimization can be applied in a linear, non-linear and integer linear model. Nonetheless, the objective function of the optimization model is of a greater significance to be considered. More so, for reasons of sustainable supply chain not always having to optimize monetary values or values that may be projected to a monetary level via an intermediary step, Schreiber (2019) argued that, there has to be particular attention given to the simultaneous inclusion of both economic and environmental factors.

In line of Schreiber’s (2019) argument, Engel et al (2009) postulated that, there are four varying alternatives of the optimization model. They viewed the first as where ecological considerations are directly projected to monetary values. Here, the results from the best offered employed to achieve energy efficiency are converted into monetary values and resulting energy costs directly incorporated into the target function. A typical example is the avoidance of the British carbon tax which can lead to a reduction of the company’s expenditure when carbon emission is managed through a sustainable supply chain design model. With the optimization model, this preclusion allows a quantification and it is possible to insert it into a cost-based target function depending on the present value of an emission allowance. Engel et al (2009) posited that, a second option is to weight specific subgoals in a higher-level objective function. Individual scaling factors for economic and ecological sub-objectives are specified for this purpose. According to Rösler (2003) however, when using weighting factors, determining the appropriate scaling factors is difficult. The third alternative as mentioned by Engel et al (2009) is the establishment of ecological constraints. This involves the use of ecological factors as constraints for economic decisions. With the fourth alternative, they argued that, this can be a multi-objective optimization. This is built on the fact that, there are many factors that account for the optimization in sustainable supply chain design. It is rather unfortunate that, these target values affect each other negatively. As a result, Ehrgott (2005) affirmed that, it is problematic to find the most efficient trade-off solution that allows for losses in one or more target values. However, the adoption of a multi-objective optimization makes it possible for the derivation of the best ultimate result based on the decision-makers' individual goal preferences (Habenicht, et al., 2003).

With these four different options for optimization, there had been a lot of studies in existing literature that surround them. However, these studies from scientist like Abdallah, et al. (2012), Arslan and Turkay (2013) and Lee, et al. (2018) have shown that the application of optimization in sustainable supply chain design can be categorized into three main scopes. The first category can be said to analytically sound initial partner selection with integrated sourcing process design based on the inclusion of sustainable criteria. Under this category, it was found that most of the researchers used employed fuzzy logic linked to a subsequent linear optimization. Particularly, scientists like Torgul and Paksoy (2019) work suggested a hybrid of a fuzzy-based analytical hierarchy process (FAHP) and fuzzy TOPSIS, followed by multi-objective linear optimization. Similarly, Azadnia et al. (2015) employed fuzzy logic in the context of a linear optimization with simultaneous order quantity calculation for the orchestration of partner selection and strategic sourcing process design, while Yu and Su (2017) used fuzzy logic connected with a data envelopment analysis (Fuzzy-DEA).

The second group of literature is also focused on the establishment of the network processes depending on the lot size determination. It basically covers the main links and processes of the entire supply chain which includes sourcing, production and distribution processes. A couple of researchers have also applied this model. For instance, in Arslan and Turkay (2013), they employed an extension of the classical economic order quantity (EOQ) calculation to a sustainable economic order quantity (SEOQ). This model allowed for the integration of sustainable components using linear optimization models with ecological constraints. More so, Priyan (2019) created an extended nonlinear optimization model as a solution option. Not all, Jaber et al. (2013) calculated optimal production rates in a supplier-producer relationship using another model which can be explained by network processes.

The third category observed from existing literature in this regard had to do with the connection of the nodes and edges of the network of a supply chain. Using a mathematical model, the decision variables reflect the transit volumes of particular goods, semi-finished products, and raw materials on the network's edges. Lee et al (2018) proposed that, in this model, the goods are allocated to the locations in the network whilst Chaabane (2011) mentioned that the creation of the transportation linkages can potentially be done in the same phase by giving the decision variables with an additional index as a degree of freedom to decide the mode of transportation on an edge. The most significant output of this model as far as sustainability of the supply chain is concerned include the distance bridging of the respective edge, the loading weight and the means of transport. Essentially, these are factors that have direct effect of the ecological balance or environmental sustainability of the network which goes beyond to affect the other pillars of sustainability.

Simulation and Sustainable Supply Chain Design

Simulation on the other hand is an alternative model which serves as decision support tool in designing a sustainable supply chain. Unlike optimization which mainly uses mathematical programme and functions, simulation employs experimental approach on the basis of an old created model of an already existing system (Schreiber, 2018). VDI (2014) agreed that simulation evaluates a previous model's performance, and the findings are projected onto the real system to aid decision-making. Applying simulation in sustainable supply chain design has always involved the replication of the distinct and changing behaviours of the various locations involved in the process as well as the processes on a computer model. This serves as an easy interactive tool to evaluate, monitor and analyse the various processes and the nature of locations. As such, how best the ecological impact of the supply network can be reduced by altering the entire processes and linkages become easy to do and evaluate. In the view of Rose and März (2011), the simulation progresses as a result of the occurrence of events. For example, in supply chain simulation, these events indicate the arrival of a product in a warehouse or the completion of the loading of a mode of transportation. This has been very beneficial to companies both economically and ecologically. This has caused its relevance to be highly attached to sustainable supply chain design. There has been a number of studies, with the first been recognized in 1996.

In 1996, Hirsch, et al. (1996) came up with one of the early simulative models. In order to facilitate the construction of environmentally friendly value networks, associated commercially customary goals such as service level and delivery time were simulatively assessed by Hirsch et al (1996) in terms of their ecological compatibility. Some years forward, Reeker et al. (2011) found appropriate methodologies for logistical expenditure and performance analysis, as well as ecological assessment, and incorporated them into an integrative evaluation methodology based on key indicators. This led to a simulation model which demonstrated the interdependence of the individual goal values inside the evaluation process. Similarly, Cirullies (2016) enhanced the logistics goal system to incorporate eco-indicators such as resource usage, emission values, and energy consumption. This development has grown over the years into a computer based simulation programmes which makes it very easy to simulate the global supply chain.

Recognizing the differences and similarities between the two approaches, it is important to note that they can be linked to achieve a sustainable supply chain design as argued by März et al (2011) in the diagram below.

Case Study

In pursuit of reducing emission of carbon in the European continent, there has been the institutionalization of policies like the European Union Emissions Trading System (EU-ETS). Despite this being an environmental sustainability legislation, it has proven to have a rippling effect of the supply chain of companies. This led to the development of an optimization model by Jaber et al (2013) to compute the optimum production rate a company can choose considering the existing conditions of carbon tax and carbon emission penalty.

Model of the Case Study

In this study, Jabers et al (2013) took into account carbon trading and established a mathematical programming problem for a two-level supply chain (vendor–buyer). The vendor's production rate was presumed to be a function of greenhouse gas emissions. The developed model's goal was to determine the optimal production rate (and thus the joint lot sizing policy) that minimizes total supply chain costs while accounting for emissions certification limits, penalties for exceeding emissions quotas, and capital invested to increase emissions limits by purchasing new certificates.

Essentially, this model predicts that incorporating carbon emissions taxes and allowances in the total cost function allows the estimation of influence of the EU and US greenhouse gas emissions reduction programs on the overall cost of operating a supply chain as well as the managerial decisions made by supply chain partners. Furthermore, by merging both techniques, we can examine the impact of a third greenhouse gas emission reduction scheme that combines emissions allowances and emissions taxes.

Findings of Case Study

Jaber et al (2013) made some intriguing observations from the developed model after subjecting 6 real life scenarios into the model. The model clearly demonstrated the behavior of the supply chain cost function for a variety of situations that may depict various legislative systems. It was discovered that a policy that considers a penalty for emissions merely provides the supply chain decision-maker with more than one choice, which may result in an optimal solution that suggests creating excessive levels of greenhouse gas emissions.

Conversely, a policy combining a carbon tax and an emissions penalty was shown to be the most successful, since the best solution developed was frequently connected with reduced emissions. When emissions and penalty costs were evaluated, supply chain coordination was shown to reduce total system costs; however, the decrease was in inventory-related expenses with no reduction in total system costs.

Overtly, this finding which proves that adopting a multi-objective approach of considering both carbon tax and emission penalty into the optimization model can be inferred to be a pareto model which was opined by Engel et al (2009) to be the best. The finding therefore confirms this argument.

Conclusion and Findings

There has been considerable number of studies done on the application of both optimization and simulation models for the design of sustainable supply chain. This started in the late 20^th century with studies like Hirsch, et al. (1996). However, these models became predominantly in the 21^st century which have attracted many academic writings. Nonetheless, one of the models have been given much attention than the other adjudging from the available academic literature found from the review. The optimization model happens to be much researched about with regards to its application to sustainable supply chain design, hence much more applied than the simulation model. From observations made, despite the relevance of simulation model, it depends heavily on previous or existing model and this cannot be possible when there is no existing model. However, the flexibility, manipulative and computational nature of the optimization model makes it easy to work with.

It was found that, each model or approach come with unique characteristics and relevance that should not be ignored in our quest to design a sustainable supply chain. This suggests that, a single approach application may limit the advantages of the other approach. In essence, there is a need to find a holistic approach of integrating both models in our efforts of predicting the most efficient way which will make the supply chain sustainable. I will however agree with Schreiber (2018) that, a holistic approach is the way to go and not a single approach. It is also crucial to note that, in the adoption of a holistic approach, there must be a consideration of every necessary supply chain design activities that ensure adequate integration of environmental characteristics in all components based on a superordinate individual key performance indicator system. More so, there must be a feedback mechanism that helps to evaluate the performance of the model. Lastly, this should be ensured to be able to be repeated as many times as needed, providing a gradually better setup based on the user's preferences.

 

 

Posted by,

Lokesh Venkatachalam