- Synthesis of Regional Renewable Supply Networks
- Introduction
- Concept and Approach for the Synthesis of Truly Sustainable Renewable Supply Networks
- Concept of system-wide supply networks and a multi-layer superstructure
- Concept of sustainability net present value
- Derivation of sustainability net present value
- Truly sustainable solutions
- General mathematical programming formulation

# Synthesis of Regional Renewable Supply Networks

## Introduction

Traditionally, production systems follow linear supply chains, from extraction, through (pre-)processing, to use and final disposal (Korliouen et al., 2018). However, such linear patterns of production and consumption are not sustainable and lead to resource depletion and environmental degradation (Merli et ah, 2018). Circular economy provides an alternative that is cyclical and offers the potential for progress towards sustainable development (Schroeder et ah, 2019). Circular economy, with its cradle-to-cradle (also closed-loop or zero-waste) philosophy, focuses on optimizing environmental unburdening (“doing more good”), as compared to optimizing environmental burdening (“doing less harm”) (Toxopeus et ah, 2015). It offers an opportunity for sustainable production and consumption based on limitless resources and continuous growth (Govindan and Hasauagic, 2018) and is closely connected to resource sustainability, emission reduction and waste management (Koh et ah, 2017).

Circular economy has recently gamed considerable attention in industry (Niero et ah, 2018), from several governments (Korhonen et ah, 2018), researchers and others (Merli et ah, 2018). It could be viewed as the integration of self-sustaining production supply networks (Genovese et ah, 2017) and sustainability, with concomitant closing of material and energy loops. It is interlinked with the industrial ecology concept, which aims at a continuous exchange of energy and materials within production systems in a sustainable way (Arbolino et ah, 2018), with the ultimate goal of achieving zero impact on the environment (Leigh and Li, 2015). A suitable means of designing closed-loop or circular production systems is the concept of sustainable supply chain networks (Winkler, 2011).

Sustainable supply chain networks need to balance economic, environmental and social performance. However, views on what constitutes “sustainable”, differ at various levels, at the levels of individuals, companies, local communities, governments and globally, owing to differences in values, interests, contexts (Mascarenhas et ah, 2014) and diversities. Sustainability is, thus, defined and assessed differently at the micro (smaller-scale) and macro (larger-scale) levels (Zore et ah, 2016). It is important that a priority for supply network designs should be to create them to be truly sustainable and not just less unsustainable

(Pagell and Shevchenko, 2014). A truly sustainable supply chain network exhibits at least non-negative economic performance, while doing at least no harm to environment and society (Pagell and Shevchenko, 2014). Synthesis of more sustainable and especially truly sustainable systems is, thus, a complex multiobjective problem (Azapagic et al., 2016). It is also important, that sustainability of systems be evaluated from a longer term perspective, across an entire system’s lifetime (Zore et ah, 2018b).

In this Chapter, the methodology for the synthesis of truly sustainable system-wide supply networks is introduced by using a mathematical programming approach, with the objective of maximizing Sustainability Net Present Value. Furthermore, the proposed methodology is demonstrated on the synthesis of larger, subcontinental-scale supply networks of Central Europe, producing food, fuel and renewable electricity. The Central Europe-“subcontinental” scale case study demonstrates partial transition from the current, mainly fossil-based energy supply, to renewable energy supply over a tune horizon of 20 years (between 2020 and 2040).

## Concept and Approach for the Synthesis of Truly Sustainable Renewable Supply Networks

In this section, a methodology for truly sustainable synthesis of renewable-based supply networks for producing energy and bioproducts is described. It is based on (i) a concept of System-Wide Supply Networks (SWSN) and the related multi-layer superstructure, (ii) a concept of Sustainability Net Present Value, and (iii) the use of a mathematical programming approach.

### Concept of system-wide supply networks and a multi-layer superstructure

A system-wide supply network is a network comprising multiple supply networks, e.g., agricultural, chemical, biochemical and energy supply networks for shared production of food, fuel and electricity. It can represent fossil- and/or renewable-based production spread locally, regionally or even globally; for example. Figure 1 illustrates a continental (European) SWSN. Each constituent supply network is further composed of separate elements. The synthesis of (bio)cliemical supply networks deals with the integration of layers from molecules and process units, to production plants and their networks. In the case of food and biomass supply and demand networks, we deal with planting and harvesting, collecting and pre-processing, processing in biorefineries, and distributing products to end users. Similarly, the energy supply network is composed of a basic layer where energy is released by means of chemical, nuclear or other physical phenomena; the next layers deal with conversion and the last one with energy distribution and use. Each of these supply networks is accompanied by a suitable superstructure; for example. Figure 2 shows a four-layer superstructure for a larger-scale biomass supply and demand network where the arrows represent the flows *q* between and within the layers. For more details regarding the four-layer superstructure and its mathematical formulation the readers are referred to the paper by Cucek et al. (2014a). Note that in a SWSN, all these four layers belonging to different supply networks are merged into an overall network and constitute an overall superstructure.

### Concept of sustainability net present value

Generally, the synthesis of sustainable systems is a complex, multi-criteria problem, since it should offer good solutions from all three sustainability pillars: Economic, environmental and societal. Research studies har e shown that different criteria pror ide different “optimal” solutions (Novak Pintaric and Kravanja, 2015). Multi-criteria optimization is of special importance if we want to achieve the best sustainable solutions. Among the methods that enable simultaneous optimization of sustainability indicators (economic, environmental and social) are the methods called Sustainability Profit (SP) (Zore et al., 2017) and Sustainability Net Present Value (SNPV) (Zore et al.. 2018b) from different micro- and macroeconomic perspectives (Zore et al., 2018a), which are defined on a monetary basis and.

Figure 1. Continental system-wide supply network composing agricultural, (bio)chemical and energy networks (after Zore et al., 2018b, based on representation of chemical supply chains by Marquardt et al., 1999).

thus, enable simple merging of all three indicators. In the case of SNPV, it is composed of Economic (A'PI^{/Eco}“^{onuc}), Eco (A'PI^{/Eco}), and Social *(NP*F^{50}™^{1}) net present rvalues, see equation (1):

This economic-based method helps to avoid the use of weighting, normalization and multidimensionality of the problem and is easy to interpret and understand. Since separate criteria are now merged, a multi-objective optimization problem is reduced to a single-objective one, which is particularly convenient for the synthesis of larger scale systems—and regional renewable supply networks are among such larger scale systems. Note that SNPV is an extension of SP, since the annual cash flows used to calculate NPV are almost identical to SP. Although SNPV is somewhat more difficult to calculate, it possesses two important advantages over SP; it considers the time value of money and is suitable for sustainability assessment over a system’s entire lifetime, which, for strategic long-term investment planning, enables one to obtain appropriate trade-off solutions between economic efficiency, environmental (un)burdening and social responsibility. Note that, besides SNPV, which is located at the intersection of all three sustainability pillars, there are three different binary combinations: Viability NPV as Economic plus Eco NPVs, Equitability NPV as Economic plus Social NPVs, and Bearability NPV as Eco plus Social NPVs—see Figure 3. We can calculate SNPV from different viewpoints: Micro- (industrial), macro- (industrial plus governmental), and wider macro-economic (industrial, plus governmental, plus employees). Since sustainable development should be directed towards improving general welfare, a wider macro-economic view has been considered for SNPV calculation.

**Figure 2. **Four-layer superstructure for biomass supply and demand networks (after Cucek et al., 2014a, adapted from Lam

et al., 2011).

**Figure 3. **Sustainability net present value is located at the intersection of all three sustainability pillars (from Zore et al., 2018b, based on a representation of sustainability by Dreo, 2016).

#### Derivation of sustainability net present value

**Note that SNPV is defined as incremental SNPV (ASNPV), i.e., as the difference between the new and old system’s SNPVs, where each NPV is calculated over lifetime Г as a sum of the corresponding incremental net annual cash flows {AFC), each one being discounted from its time t to the present time (t = 0) with a discount rate of r _{d}.**

**2.2.1.1 Incremental Economic NPV**

This is defined as the difference between the Economic NPV of new alternatives and the Economic NPV of old alternatives. For the purpose of simplicity, we assume that old alternatives operate close to zero NPV. This assumption allows us to define incremental economic net annual cash flow only in terms of new alternatives. From micro-economic or industrial perspective, it is defined as the after-tax surplus of revenue *(R)* and additional revenue from governmental subsidies or higher redemption prices *(AR)* over expenditures *(E),* eco tax (E™ “*) and annual depreciation (£>^{Economlc}); plus /)^{Ес01К>т1С}. minus investments (^Economic) pj_{us} p_{0SS}ibi_{e} salvage value (SF^^{000}™'), all defined at year *t.*

Incremental Economic NPV from an industrial perspective is then defined:

From the wider macro-economic perspective (industrial, plus governmental, plus employees), the definition of net annual cash flow becomes much simpler, because taxes paid by companies and governmental incomes from these taxes are cancelled out.

Note that net annual cash flow for employees is omitted, since it is considered in Social NPV. Incremental Economic NPV from the wider macro-economic perspective is finally defined:

**2.2.1.2 Incremental Eco NPV**

It is defined as the difference between the Eco NPV of new alternatives and the Eco NPV of old ones, where eco annual cash flows for both new and old alternatives are calculated using eco-cost coefficients (Delft University of Technology, 2019). Eco-cost coefficients represent marginal costs that one would have to spend per product unit in order to prevent environmental burdening. The eco-cost of a company *(EC),* related to the use of resources and production of products at year *t,* can be calculated as the stun of mass and energy flows multiplied by the corresponding eco-cost coefficients. Note that net eco-cost related to the transition from old, environmentally harmful production alternatives to greener ones is usually negative because the eco-cost related to new alternatives *(ECf")* is usually smaller compared to that for old alternatives (£C°^{ld}). By shutting down old alternatives, charges related to their eco-cost are released and saved. Their £C^{old} thus represents the eco benefit *(EB).* The difference between *EB _{t}* and the eco-cost related to new alternatives (£C“'") is the eco-profit

*(EP)*which, in the definition of Eco NPV, is equal to its incremental eco net annual cash flowJ£C

^{Ec0}. If production alternatives need subsidies or higher redemption product prices

*(AR)*in order to operate with a positive economic NPV, this extra money resembles additional eco-cost charged in order to avoid burdening the environment. Incremental eco net annual cash flow is then defined as:

and incremental Eco NPV as:

**2.2.1.3 Incremental Social NPV**

Similarly, this is the surplus of the Social NPV of new alternatives over the Social NPV of old ones. Its incremental net annual cash flow is defined as a function of a net number of jobs (new jobs created, minus old jobs lost), _TV^{)0b} at year *t:*

where terms represent:

- • JSS: Social security contributions to the state paid by employees and employers defined as the difference between average gross and net salaries in the production sector, multiplied by
*AA*^{T}>^{ob} - •
*JSU*Social unburdening of the state budget because of new job creation defined as the product of_{r}:*AN}°*and the average state social transfer for unemployed people,^{bs} - •
*ASC*Social cost of the state and the company for employees defined as the product of_{r}:*АА*and the sum of an average state social transfer and an average company’s social charge per employee,^{Т}}°^{Ы} - • J.S’.8
^{Employet!}: Social benefit from the employees prospective corresponding to net salary plus social benefits provided to a net number (JV_{f}^{)obs}) of employees by the state and companies, and - • riSCj
^{EmployM!}: From the newly employed employees prospective, this term represents the loss of their unemployment support from the government defined as the product of JjV^{)ob5}and the average state social transfer for unemployed people.

Note that, because from the wider macro-economic perspective most of the parts in the above terms cancel each other out, the incremental net annual cash flow simply becomes equal to the stun of gross salaries, AGS, defined as the product of J,VJ^{obs}and the average employee’s gross salary.

Incremental Social NPV is then defined as:

**2.2.1.4 Incremental Sustainability NPV**

SNPV comprising Economic, Eco and Social NPVs is finally defined:

Note that term *AR _{t}* is cancelled out. For more details, readers are referred to the papers by Zore et al. (2017, 2018a, b). For simplicity, the symbol “Д” will be omitted in the continuation.

#### Truly sustainable solutions

As a very rough condition, it can be stated that a solution is sustainable if it has positive or at least nonnegative SNPV. However, it may happen that even when a solution possesses very high value of SNPV, one of its NPVs can still be negative. For a system to be truly sustainable, a necessary condition should, therefore, be consideration of all its constituent parts. Any serious deficiency in any sustainability pillar would cause such a system to not sustain current and future challenges and, hence, fail to endure. Therefore, a genuinely sustainable solution should stay within the intersection where all three sustainability pillars are at least nonnegative:

An even stricter condition is that any solution obtained must also be economically feasible from the industrial point of view. A uounegative constraint for economic NPV from the industrial perspective, see equation (13), should be imposed on a group of selected production plants as a whole:

Note that the above constraints form a feasible region for a family of genuinely sustainable solutions where the most sustainable is, in principle, the one that is obtained by maximizing SNPY subjected to the above nonnegativity constraints:

Note also that once maximal SNPV was obtained with its resulting NPVs being maximal, ,ypf/E^{c}_{s} p_{0SS}ible to obtain any neighboring solution by imposing

the following additional constraints:

where *d ^{a},b^{b}* and

*d*are the corresponding distances, in either positive or negative directions.

^{c}### General mathematical programming formulation

Based on the above concept of Sustainability Net Present Value, a general mixed-integer (non)linear programming (MI(N)LP) formulation for a System-Wide Supply Network is posed and the (SWSN- MI(N)LP) problem is presented below:

Different objectives can be composed by setting suitable 0-1 values to the weightings u’",-u^{i>} and v/. Wien all values are 1, the objective function (a) would resemble the SNPV to be maximized. Equality constraints (b) usually represent mass and energy balances, and design equations. Inequality constraints (c) denote product quantity and quality specifications, operational, environmental, and feasibility constraints. Logical constraints (d) are used for the selection/rejectiou of sustainable alternatives, e.g., routes, zones, processes, etc. The remaining inequality constraints (e-h) represent nonnegativity sustainability constraints. Note that constraints (b-d) are defined at different levels *I* 6 *L,* for different production technologies *p* 6 *P,* across several supply networks, 5 E 5, at time periods *1*6 *T* over the whole system’s lifetime. Variables x in equation (16) represent continuous variables and у binary variables/ decisions related to topology of supply network.

The following section presents an application of the above methodology to the synthesis of a supply network producing food, fuel and electricity for targeting the partial transition from fossil-based to renewable resources on a subcontinental scale of Central Europe over a tune horizon of 20 years.