sustainability

Article Using Specification and Description Language for Life Cycle Assesment in Buildings

Pau Fonseca i Casas 1,* and Antoni Fonseca i Casas 2

1 Statistics and Operations Department, Universitat Politècnica de Catalunya, Barcelona-Tech, 08034 Barcelona, Spain 2 Polyhedra Tech SL, 08026 Barcelona, Spain; [email protected] * Correspondence: [email protected]; Tel.: +34-934-017-732

Academic Editor: Avi Friedman Received: 10 March 2017; Accepted: 5 June 2017; Published: 10 June 2017

Abstract: The definition of a Life Cycle Assesment (LCA) for a building or an urban area is a complex task due to the inherent complexity of all the elements that must be considered. Furthermore, a multidisciplinary approach is required due to the different sources of knowledge involved in this project. This multidisciplinary approach makes it necessary to use formal language to fully represent the complexity of the used models. In this paper, we explore the use of Specification and Description Language (SDL) to represent the LCA of a building and residential area. We also introduce a tool that uses this idea to implement an optimization and simulation mechanism to define the optimal solution for the sustainability of a specific building or residential.

Keywords: Specification and Description Language (SDL); Life Cycle Assesment (LCA); smart cities; IoT; buildings simulation; sustainability; resilience

1. Introduction Currently, there is worldwide awareness of the global increase in energy consumption and the limited amount of resources available to provide energy to citizens and enterprises. This creates a huge energy demand and the need to define a new economy and transition methods to increase energy savings [1]. Considering more effective methods to reduce energy consumption in residential and commercial areas is strongly related to building efficiency, which has a great impact on electricity consumption. Although electricity represents only a part of the total energy consumption, it clearly reflects the importance of the building in the global impact. In the USA, the residential sector represents approximately 38% of the electricity consumed and in the commercial area, approximately 36% [2]. The Life Cycle Assesment (LCA) in a building or an urban area helps to obtain a clear image of all processes involved in the planning, construction, use, and deconstruction of a house. This holistic view can assist in the reduction of the impact, taking care of the economic, social, and environmental aspects in a cradle to cradle approach [3]. From an environmental point of view, reducing energy consumption would have a great impact in decreasing the environmental impacts. In addition, the analysis of the LCA for houses has a great impact on energy consumption in all sectors, not only in houses, but also shows a foreseeable greater impact on the residential and commercial sectors that represent a huge part of the total. For example, all materials that must be used in a building must be transported from several different origins to its final destination of the building. By incorporating these parameters in the model, we can also reduce the emissions from transportation, which are huge.

Sustainability 2017, 9, 1004; doi:10.3390/su9061004 www.mdpi.com/journal/sustainability Sustainability 2017, 9, 1004 2 of 17

2. State of the Art Building or residential area LCA is complex, mainly because it needs to use powerful tools to scope the complex dynamics of the environment and the energy. Several problems arise with this type of analysis and includes the amount and the diversity of the data that must be used, and the fact that this area utilizes different specialists working with different skills in multidisciplinary teams. Mainly, these problems are related to the choice of unit of the analysis and the methodological and practical approaches used [4]. In this context, some approaches have been undertaken to improve the LCA application in architecture [5,6]. A complete review of the development of the LCA in the field of architecture can be found in the studies presented by [7,8]. By 31 December 2020, EU directive 2010/31/EU requires that all new buildings must be nearly zero-energy buildings (nZEB). The aim of this directive is to incorporate concepts for reducing their impact on the environment [9]; thus, the member states must also complete the thresholds and policies to update their building stock [10–13]. Member states have launched a series of projects, actions, and policy guidelines to specify the optimal cost-performance actions to achieve the objectives proposed by the regulation on energy rehabilitation (Royal Decree 235/2013). This allows researchers to check which urban areas are in the greatest need of rehabilitation and/or that suffer from so-called “energy poverty” [14]. Furthermore, several European and national projects have been proposed [15,16] to provide funding for initiatives which, in addition to addressing the issue of energy efficiency either for new construction or rehabilitation, also include the impact of construction on the environment; that is to say, be capable of conducting a LCA (Life Cycle Assesment) of the building and thereby ascertain its associated impacts. Such concepts must be addressed to complete the analysis, not solely in terms of energy and economic cost. To perform a comprehensive study of sustainability, one must include the proposed European standards developed by the CEN TC 350 Working Group, and implemented nationally via Standard UNE EN 15643 on the Sustainability of Construction Works [17]. This set of standards describes and specifies the impacts that must be considered (environmental, economic and social impacts), details indicators, and proposes methodologies. If we consider the criteria and objectives that must be achieved based on the above-mentioned standards, programs and applications must be employed to simulate energy consumption, calculate economic and environmental impacts, and conduct a complete LCA. No global proposals exist for analyzing certain parameters. Often the designer or engineer has to invest a significant amount of time designing a virtual model just to study a few design iterations to identify a good solution. The designer is not in a position to ensure that the best solution is adopted, or at least one of the best solutions, without analyzing all, or an accurately selected subset, of the possibilities. In addition, most commercial simulation software packages allow for a building’s annual energy analysis to be calculated, but do not perform any kind of optimization for the process. Hence, none of the results for the previous steps, such as the design, construction, or demolition processes, are optimized. Efforts have been made to develop approaches and methods for assessing buildings [18], methods for integrating daylight and thermal efficiency [19–21], photovoltaic systems [22], and multi-objective optimization systems into urban design [23], which consider several aspects when creating an energy optimization system. Languages that are commonly used for programming simulators or optimization tools are C++, C#, Fortran, Java and Delphi, among many others. The use of different languages may cause problems with integration, model determination and understanding between researchers from different disciplines, as not everyone understands the low-level code finally used. One of the proposals presented in this work is the use of a formal language to simplify model definition and use. The use of formal languages, such as Discrete Event System Specification (DEVS) [24], Specification and Description Language (SDL) [25,26], or Petri nets [27], is without a doubt one of the best solutions. These languages allow for an easy integration and communication of ideas related to Sustainability 2017, 9, 1004 3 of 17 the model with the other members of the team. Additionally, these tools improve interoperability with other models (co-simulation) and promotes cooperation. Examples of recent works that explore the use of official languages in this area are [28,29], which explore the use of the co-simulation approach for calculating building HVAC systems (heating, ventilation, and air conditioning). To represent this LCA, we used a discrete simulation model formalized using Specification and Description Language (SDL). In this area, several approaches exist, mainly ruled by the tools used. Classical languages like GPSS/H® [30], Slam II® [31], or others considered more modern like Simprocess® [32], Arena [33], or SIMIO [34], among many others, can be used. These software packages allow a complete definition of the model behavior, structure, time and representation; however, once the model is defined in one of these simulation packages, it is often difficult to migrate to another. All specialists who interact with the model must also understand how the tool works. In this context, the use of formal language to represent the model and to later use a specific tool to implement the model is necessary, even more so if the model is complex or the personnel working on the model own different forms or, do not know how to implement the model in the specific tool selected to perform its implementation. Programming libraries and infrastructures exist that allow for the definition of a simulation model following a formal language. The tools related to DEVS [35–37] and PetriNets formalisms [27] often represent an excellent alternative to define and implement a simulation model based on Petri nets or DEVS. Specifically to model buildings and their related processes, several experiences exist using DEVS as a language, based on the BIM (Building Information Modeling) philosophy, see [28,38], but no experiences trying to model the entire LCA of a house or urban area have been found. SDL is not new for modeling energy-related issues. [39] presented two complementary approaches to specify energy aspects during the design phase of wireless networks, but like DEVS, no studies that model the LCA of a building exist. In this paper, we describe our experience of using SDL to describe the main processes that define the LCA of a building and show how to take advantage of a formal representation to expand the model implementation easily over different platforms. The paper is organized as follows: in Section3, we describe the SDL based model we use; in Section4, we discuss how to implement this model; in Section5, we describe how to execute the model on the web using a cloud infrastructure. Finally, Section6 contains our conclusions.

3. Methodology One of the aims of this research was to describe, as formally as possible, the structure and behavior of the LCA of a building. This goal was crucial as the people involved are diverse and come from many different backgrounds. To address this issue, we used SDL to formally define the model. The use of the SDL language afforded maximum flexibility to integrate new processes and procedures to the system in a co-simulation scenario [29]. Using a graphical formal language allowed nonspecialized technicians access to the programming details of the model and definitively contributed to the validation and verification processes. In addition, the explanation of the model itself was more intuitive, simple, and direct than using a programming language such as C or Java. Thus, the group can share the model, making it much easier to understand and detect errors, as well as propose improvements.

3.1. Specification and Description Language In this section, we do not describe specifically how SDL behaves, more information on this language can be found in [25,26]. However, we want to note that the graphical capabilities, the completeness, and lack of language ambiguity makes it a great candidate to formally define any kind of simulation model. Specification and Description Language (SDL) is an object-oriented, formal language defined by The International Telecommunications Union-Telecommunications Standardization Sector [25], formerly Comité Consultatif International Telegraphique et Telephonique (CCITT), in recommendation Z.100. The language is intended for the specification of complex, event-driven, real-time, and Sustainability 2017, 9, 1004 4 of 17 Sustainability 2017,, 9,, 1004 4 of 17 SustainabilitySustainability 2017 2017, ,9 9, ,1 1004004 44 of of 17 17 SustainabilitySustainability 2017 2017, 9, 9, 1, 1004004 44 of of 17 17 discreteinteractive signals. applications SDL uses involving four levels many to concurrent describe model activities behaviors that communicate:: system,system, blocks,blocks, using processes, discreteprocesses, signals. andand discrete signals. SDL uses four levels to describe model behaviors:: system,system, blocks,blocks, processes,processes, andand discreteproceduresSDL uses signals. four,, seesee levels FigureFigureSDL uses to 11.. describe four levels model to behaviors:describe model system, behaviors blocks,: system, processes, blocks, and procedures,processes, and see proceduresprocedures, ,see see Figure Figure 1 1. . proceduresFigureprocedures1. , ,see see Figure Figure 1 1. .

Figure 1.. Structural view of a Specification and Description Language (SDL)(SDL) System. FigureFigure 1 1.1. .Structural. Structural view view of of a a S SpecificationSpecificationpecification and andand D DescriptionDescriptionescription L LanguageLanguageanguage (SDL) (SDL)(SDL)(SDL) System. System.System. With SDL,, itit isis possiblepossible toto generategenerate unambiguousunambiguous codecode fromfrom thethe specification;specification; thus,thus, thethe idealideal toto WithWithWith SDL SDL,SDL, ,,it ititit is isisis possible possiblepossiblepossible to tototo generate generategenerategenerate unambiguous unambiguousunambiguousunambiguous code codecode from fromfrom the thethe specification; specification;specification;specification; thus, thus, the the ideal ideal to to avoid the implementationimplementation process process and and assure assure that that the the current current implementation implementation of of the the model model avoidavoidavoid the the the implementation implementationimplementationimplementation process process process process and and assureand and assure assure assure that the that that that current the the the current implementation current current implementation implementation implementation of the model of of of represented, the the the model model model represented,, asas closelyclosely asas possible,possible, thethe modelmodel proposed.proposed. SDL has different advantages: it is a non-- representasrepresent closelyeded as, ,,as as possible,as closely closelyclosely as the asas possible, possible,possible, model proposed. the thethe model modelmodel SDLproposed. proposed.proposed. has different SDL SDL has has advantages: different different advantages: advantages: it is a non-ambiguous it it is is a a non non-- ambiguous language, graphical, and standardized (the ITU--T), and as SDL allows thethe definitiondefinition ofof ambiguouslanguage,ambiguous graphical,language, language, andgraphical, graphical, standardized and and standardized standardized (the ITU-T), (the (the and ITU ITU as- SDLT),-T), and and allows as as SDL SDL the definitionallows allows the the ofdefinition definition distributed of of distributed systems, the resulting model can be executed over different computers without any distributedsystems,distributed the systems, systems, resulting the the model resulting resulting can be model model executed can can over be be executedexecuted different over computers over different different without computers computers any modification without without any any of modification of the model representation. This allows a process where a model can be implementimplemented modificationthemodification model representation. of of the the model model representation. representation. This allows a processThis This allows allows where a a process process a model where where can be a a model implementedmodel can can be be implement andimplementimplement executededed and executed inin distributed,distributed, parallelparallel,, or sequential ways without any modification; the only issue is andinand distributed, execute executedd in inin parallel, distributed, distributed,distributed, or sequential parallel parallelparallel, ,,or waysor sequ sequ withoutentialential ways ways any without modification; without any any modification; modification; the only issue the the is only selectingonly issue issue theis is selecting the proper tool to perform thethe execution or implementation. SDL is designed to assure its selectingproperselecting tool the the toproper proper perform tool tool the to to executionperform perform the the or execution implementation. execution or or implementation. implementation. SDL is designed SDL SDL to is is assuredesigned designed its compatibilityto to assure assure its its compatibility with Unified Modeling Language (UML) and Message Sequence Chart (MSC)(MSC) (see(see compatibilitywithcompatibility Unified Modeling with with UnifiedUnified Language Modeling Modeling (UML) Language Language and Message (UML) (UML) Sequence andand MessageMessage Chart (MSC) Sequence Sequence (see Chartdocument Chart (MSC) (MSC) Z.105 (see(see of document Z.105 of the ITU that describes how to use Message Sequence Charts in combination with documentthedocument ITU that Z. Z.105 describes105 of of the the howITU ITU that to that use describes describes Message how how Sequence to to use use Charts Message Message in combinationSequence Sequence Charts Charts with in SDL). in combination combination SDL, like UML,with with SDL). SDL, like UML, is a standard of the ITU, and one of the main concerns is assuring compatibility SDL).isSDL). a standard SDL, SDL, like like of UML, theUML, ITU, is is a a andstandard standard one of of of the the the main ITU, ITU, concernsand and one one of is of assuringthe the main main compatibilityconcerns concerns is is assuring assuring between compatibility compatibility the different between the different proposed standards. As SDL is compatible with some extensions thatt are are betweenproposedbetween the the standards. different different As proposed proposed SDL is compatible standards. standards. with AsAs SDL some SDL is isextensions compatible compatible that with with are sproposed someome extensions extensions to UML, tha tha sucht t are are as proposed to UML, such as SysML or AUML, that extends thethe languagelanguage tto represent intelligent agents. proposedSysMLproposed or to to AUML, UML, UML, such that such extendsas as S SysMLysML the or or language AUML, AUML, that tothat represent extend extendss the intelligentthe language language agents. t oto represent represent intelligent intelligent agents. agents. The different model concepts that the SDL formalism describes are (i)(i) Structure: system, blocks, TheTheThe different differentdifferent model modelmodel concepts conceptsconcepts that thatthat the thethe SDL SDLSDL formalism formalismformalism describes describes are are (i) (i) Structure: Structure: system, system, blocks, blocks, processes and processes hierarchy; (ii)(ii) Communication: signals, along with the parameters and processesprocessesprocesses and and and processes processes processes hierarchy; hierarchy; hierarchy; (ii) (ii)(ii) Communication: Communication:Communication: signals, signals, along along with with the the the parameters parameters parameters and and and channels that the signals use to travel; (iii)(iii) Behavior: defined through the different processes; (iv)(iv) channelschannelschannels that thatthat the the signalssignals signals useuse use to to to travel; travel; travel; (iii) (iii) (iii) Behavior: Behavior: Behavior: defined defined defined through through through the the differentthe different different processes; processes; processes; (iv) Data:(iv) (iv) Data: based on Abstract Data Types (ADT); and (v)(v) Inheritances:Inheritances: to to describe describe the the relationships relationships Data:basedData: based on based Abstract on on Abstract Abstract Data Types Data Data (ADT); Types Types and (ADT); (ADT); (v) Inheritances: andand (v)(v) Inheritances:Inheritances: to describe tothe to describerelationships describe the the relationships between, relationships and between, and specialization of, the model elements. between,specializationbetween, and and specialization specialization of, the model of, elements.of, the the model model elements. elements. System diagrams, i.e., block diagrams describing the model structure, represent hierarchical SystemSystemSystem diagrams, diagrams, diagrams, i.e., i.e., i.e., block block block diagrams diagrams diagrams describing describing describing the the the model model model structure, structure, structure, represent represent represent hierarchical hierarchical hierarchical decompositions of the different model elements; some good examples can be reviewed in [26][26].. AA decompositionsdecompositionsdecompositions of of the the the different different different model model model elements; elements; elements; some some some good good good examples examples examples can can can be be reviewed bereviewed reviewed in in [26] in[26] [. 26..A AA]. process diagram defines the behavior of the agents when a specific signal is received and uses processAprocess process diagram diagram diagram definesdefines defines the the the behavior behavior behavior of of of the the the agents agents agents when when when a a a specific specific specific signal signal signal is is is received received and and uses usesuses different graphical elements to represent its behavior. differentdifferentdifferent graphical graphicalgraphical elem elementselementsents to toto represent representrepresent its itsits behavior. behavior.behavior. Start. This element defines the initial condition for a PROCESS diagram. Start.Start.Start. This This This element element element defines defines defines the the the initial initial initial condition c conditionondition for for for a a PROCESSa PROCESS PROCESS diagram. diagram. diagram. State.State. TheThe state state element element contains contains the the name name of of a a state. state. This This element element defines defines the the states states of of State.State. The The state state element element contains contains the the name name of of a a state. state. This This element element defines defines the the states states of of behavioralbehavioral diagrams, diagrams, such such as as PROCESS PROCESS diagrams. diagrams. behavioralbehavioral diagrams, diagrams, such such as as PROCESS PROCESS diagrams. diagrams. Input.Input. InputInput elements elements describe describe the the type type of ofevents events that that can can be received be received by the by process. the process. All Input.Input. Input Input elements elements describe describe the the type type of of events events that that can can be be received received by by the the process. process. All All branchesAll branches of a specific of a specific state state start startwith withan InputInput anInput element element because because an object an objectonly changes only changes its state its when state branchesbranches of of a a specific specific state state start start with with an an Input Input element element because because an an object object only only changes changes its its state state when when awhen new event a new is event received. is received. aa new new event event is is received. received. Create.Create. ThisThis elementelement allowsallows thethe creationcreation ofof anan agent.agent. Create.Create. This This element element allows allows the the creation creation of of an an agent. agent. Task.Task. ThisThis element element allows allows the the interpretation interpretation of of informal informal texts texts or programmingor programming code. code. In this In Task.Task. This This element element allows allows the the interpretation interpretation of of informal informal texts texts or or programming programming code. code. In In thisthispaper, paper,paper, following followingfollowing SDL-RT SDLSDL- (PragmaDev-RT (PragmaDev(PragmaDev SARL, SARL,SARL, 2006), 2006)2006) we,,used wewe useuse C code.d C code. thisthis paper, paper, following following SDL SDL-RT-RT (PragmaDev (PragmaDev SARL, SARL, 2006) 2006), ,,we wewe use useusedd C C code. code. Procedure call. These elements perform a procedure call. A PROCEDURE can be defined ProcedureProcedure call. call. These These elements elements perform perform a a procedure procedure call. call. A A PROCEDURE PROCEDURE can can be be defined defined inin thethe lastlast levellevel ofof thethe SDLSDL languagelanguage and used to encapsulate pieces of the model for reuse. ininin the thethe last lastlast level levellevel of ofof the thethe SDL SDLSDL language languagelanguage and and used used to to encapsulate encapsulate pieces pieces of of the the model model for for reuse. reuse. Sustainability 2017, 9, 1004 4 of 17 discrete signals. SDL uses four levels to describe model behaviors: system, blocks, processes, and procedures, see Figure 1.

Figure 1. Structural view of a Specification and Description Language (SDL) System.

With SDL, it is possible to generate unambiguous code from the specification; thus, the ideal to avoid the implementation process and assure that the current implementation of the model represented, as closely as possible, the model proposed. SDL has different advantages: it is a non- ambiguous language, graphical, and standardized (the ITU-T), and as SDL allows the definition of distributed systems, the resulting model can be executed over different computers without any modification of the model representation. This allows a process where a model can be implemented and executed in distributed, parallel, or sequential ways without any modification; the only issue is selecting the proper tool to perform the execution or implementation. SDL is designed to assure its compatibility with Unified Modeling Language (UML) and Message Sequence Chart (MSC) (see document Z.105 of the ITU that describes how to use Message Sequence Charts in combination with SDL). SDL, like UML, is a standard of the ITU, and one of the main concerns is assuring compatibility between the different proposed standards. As SDL is compatible with some extensions that are proposed to UML, such as SysML or AUML, that extends the language to represent intelligent agents. The different model concepts that the SDL formalism describes are (i) Structure: system, blocks, processes and processes hierarchy; (ii) Communication: signals, along with the parameters and channels that the signals use to travel; (iii) Behavior: defined through the different processes; (iv) Data: based on Abstract Data Types (ADT); and (v) Inheritances: to describe the relationships between, and specialization of, the model elements. System diagrams, i.e., block diagrams describing the model structure, represent hierarchical decompositions of the different model elements; some good examples can be reviewed in [26]. A process diagram defines the behavior of the agents when a specific signal is received and uses different graphical elements to represent its behavior. Start. This element defines the initial condition for a PROCESS diagram. State. The state element contains the name of a state. This element defines the states of behavioral diagrams, such as PROCESS diagrams. Input. Input elements describe the type of events that can be received by the process. All branches of a specific state start with an Input element because an object only changes its state when a new event is received. Create. This element allows the creation of an agent. Sustainability 2017, 9, 1004 5 of 17 Task. This element allows the interpretation of informal texts or programming code. In Sustainability 2017, 9, 1004 5 of 17 this paper, following SDL-RT (PragmaDev SARL, 2006), we used C code.

SustainabilityProcedureProcedureOutp 2017ut. ,, 9 call.,,call. 1004 Output These Theseelements elements elements describe perform perform the types a a procedure procedure of signals call. call. to A Abe PROCEDURE PROCEDURE sent, the parameters can can be be defined definedthat5 of the 17 in inthesignal the last last carries level level of, andof the the the SDL SDL destination. language language and Ifand ambiguity used used to to encapsulate encapsulateabout the piecessignal pieces ofdestination of the the model model exists, for for reuse. reuse.communication can beOutpOutput. directedut. by Output specifying elements elements destinations describe describe using the the typesa types processing of of signals signals ident to toity be be valuesent, sent, the(PId), the parameters parameters an agent namethat that the, the or signalsignalusing carries carries,the sentence, and thethe via destination.destination. path. If there IfIf ambiguity ambiguity is more thanabout about one the the signalpath signal and destination destination no specific exists, exists, output communication communication is defined, canan canbearbitrary directedbe directed one by specifyingisby used. specifying The destinations destination destinations using value using a processing can a processing be stored identity ident in value aity variable value (PId), (PId), foran agent later an agent name, use. name Four or using , PI ord usingtheexpressions sentence the sentence can via be path. viaused If path.: there(i) self, If is there an more agent’s is than more own one than pathidentity; one and path (ii) no parent, specificand no the outputspecific agent is outputthat defined, created is defined, an the arbitrary agent an arbitraryone(Null is for used. initial one The is agents); used. destination The (iii) destinationoffspring, value can the be value most stored can recent in be a variable agent stored created infor a later variable by use.the agent; Four for laterPId and expressions use.(iv) sender, Four PI canthed expressionsbeagent used: that (i) sent can self, thebe an usedlast agent’s signal: (i) ownself, input an identity; agent’s(null before (ii) own parent, anyidentity; signal the agent(ii) rec parent,eived). that created the agent the that agent created (Null the for agent initial (Nullagents); for (iii)initial offspring, agents);the (iii) most offspring, recent the agent most created recent by agent the agent;created and by the (iv) agent; sender, and the (iv) agent sender, that sentthe Decision. These elements describe bifurcations. Their behavior depends on how the related agentthe last that signal sent inputthe last (null signal before input any (null signal before received). any signal received). question is answered. Decision.Decision.The last level These of the elements SDL languagedescribe bifurcations. (PROCEDURE Their Their diagrams) behavior behavior depends depends allows on the on how how description the the related related of questionquestion is is answered. answered. procedures that can be used in the PROCESS diagrams through the procedure calls . These TheThe last last level level of of the the SDL SDL language language (PROCEDURE (PROCEDURE diagrams) diagrams) allows the allows description the description of procedures of diagrams are very similar to the PROCESS diagrams with the exception that they do not need state proceduresthatdefinitions can be . that used can in the be PROCESS used in the diagrams PROCESS through diagrams the procedurethrough the calls procedure. These calls diagrams. These are diagramsvery similar are tovery the similar PROCESS to the diagrams PROCESS with diagrams the exception with thatthe exception they do not that need they state do definitions.not need state definitions3.2. The Model. 3.2. The Model Here, the model is described with its main elements to understand the benefits of the approach. 3.2. TheHere, Model the model is described with its main elements to understand the benefits of the approach. For a detailed description of the model please refer to [40]. Figure 2 shows the first level of the For a detailed description of the model please refer to [40]. Figure2 shows the first level of the building buildingHere, simulationthe model is model. described Four with main its main blocks elements represent to understand the environment, the benefits the of the building, approach the. simulation model. Four main blocks represent the environment, the building, the compensation and Forcompensation a detailed and description the waste of treatments the model. please refer to [40]. Figure 2 shows the first level of the buildingthe waste simulation treatments. model. Four main blocks represent the environment, the building, the compensation and the waste treatments.

FigureFigure 2 2.. SDLSDL system system diagram diagram detailing detailing the the main main components of the model considered in the study.

FigureIn this this 2 model,..model SDL system the, the structure diagram structure and detailing and behavior behavior the main of the ofcomponents building the building LCA of the areLCA model described are considered described as formally in theas formallystudy. as possible. as Thepossible. detailed The descriptiondetailed description of the model of the helps model in the helps validation in the validation process. The proce validationss. The validation was done bywas a teamdoneIn ofby th specialistsais teammodel of, onspecialiststhe the structure system, on andthe architects, system, behavior and architects of engineers the building, and who engineers clearlyLCA are understandwho described clearly the understandas system. formally This the as possible.typesystem. of validation TThehis typedetailed (forof validation validationdescription ( offor of the v thealidation conceptual model of helps the model, conceptual in the see [validation41]) model, is needed proce see to [41]ss. depict )The is needed thevalidation relationships to depict was donethe relationsby a teamhips of betweenspecialists all on th thee components system, architects that the, and experts engineers depict who as clearly important understand for system the system.behavior. This Once type this of validation has been achieved(for validation, the structure of the conceptual of the model model, is see complete [41]) is, aneedednd one to can depict start theanalyzing relations thehips model between behavior. all the components that the experts depict as important for system behavior. Once this has been achieved, the structure of the model is complete, and one can start analyzing the model behavior. Sustainability 2017, 9, 1004 6 of 17

Every building is connected to energy (isolated or in a network) and to social networking (the residents), and these concepts must be modeled to complete the cycle, an idea already introduced in Cradle to Cradle [3] where we face a new paradigm of design (see Figure 3). This vision of the reality is characterized in the SDL model presented in Figure 2. The

B1_EnvironmentSustainability 2017, 9 , 1004block encapsulates all processes related to the environment. For example,6 of 17 it represents the amount of radiation that the building receives, the hydrologic conditions, and so on. All of the information calculated in this BLOCK is sent to the B1_building block through data that representsbetween all all the of componentsthe required thatweather the expertsinformation. depict B1_Building as important is the for systemmain BLOCK behavior. of the Once model this as has it representsbeen achieved, the main the structure processes of thethat model rule the is complete,behavior and onestructure can start of the analyzing building. the B1_compensation model behavior. encapsulatesEvery building the processes is connected necessary to energy to deliver (isolated the or energy in a network) consumed and to by social the building networking and (the to neutralizeresidents), the and environmental these concepts impacts must be (through modeled the to completeprocess of the energy cycle, generation, an idea already for example, introduced wit inh solarCradle photovoltaic to Cradle [3 or] where CO2 gas we absorption face a new systems). paradigm of design (see Figure3).

Extraction of raw materials

Reuse / Transportation recycling

Use Manufacturing

Distribution

Figure 3.3. Cradle to Cradle,Cradle, seesee[ 3[3]].. Every Every house house (and (and product) product) from from the the point point of viewof view of theof the complete complete life

lifecycle. cycle The. phasesThe phases that are that now are included now included in the model in the are model Transportation, are Transportation defining a CO, defining2 consumption a CO2 consumfor the materials;ption for the Manufacturing materials; Manufacturing and Distribution, and definingDistribution, a CO defining2 consumption a CO2 consumption for the constructing for the constructingsolutions; and solutions Use, detailed; and Use, with detailed the main with processes the main of the processes model that of the represents model that the behaviorrepresents of the behaviorbuildings of during the buildings this phase. during this phase.

B1_WasteTreatment represents the processes related to waste disposal. Each of the different This vision of the reality is characterized in the SDL model presented in Figure2. blocks of the model is decomposed into a single process. The B1_Environment block encapsulates all processes related to the environment. For example, The main variables that the model uses are detailed in Table 1. The BIM model is represented by it represents the amount of radiation that the building receives, the hydrologic conditions, and the Energy+ format, idf files, and climatic information is represented on the andepw files. Several so on. All of the information calculated in this BLOCK is sent to the B1_building block through other factors (variables) can be modified in the model to define the scenarios to be executed. data that represents all of the required weather information. B1_Building is the main BLOCK of the model as it represents the main processes that rule the behavior and structure of the building. Table 1. Main input model variables. B1_compensation encapsulates the processes necessary to deliver the energy consumed by the building and to neutralizeVariable the environmental impacts (through the processDescription of energy generation, for example, This file contains the description of the climate that affects the building. In with*.epw solar file photovoltaic (climate file) or CO2 gas absorption systems). B1_WasteTreatment representsthis the case, processes the building related is located to waste in Madrid, disposal. Spain. Each of the different blocks of the*.idf model file is(model decomposed file) intoThis a single file contains process. the structure of the building (geometry, materials, etc.). Materials and constructing The main variables that theNeeded model for uses the arecalcul detailedations and in Tableis obtained1. The using BIM a modeldedicated is representeddatabase or solutions database connected to external databases like the one provided by the ITeC [42]. by the Energy+ format, idf files, and climatic information is represented on the andepw files. Several other factors (variables) can be modified in the model to define the scenarios to be executed. The main output variables are shown in Table 2 and represent only a few subsets of the information obtained from the modelTable that 1. Main is stored input in model a text variables. file following an XML format.

VariableTable 2. Main output variables. Description Variable This file contains the descriptionDescription of the climate that affects the building. *.epw file (climate file) The energy demandIn this case, of the the model building will is be located determined in Madrid, to minimize Spain. the building’s Energy Impact *.idf file (modelenergy file) (energyThis for file heating contains and the cooling). structure of the building (geometry, materials, etc.). Environmental Analysis of the environmental impact of materials used (Global Warming, Ozone Materials and constructing Needed for the calculations and is obtained using a dedicated database or Impact Depletion, etc.) according to an LCA (Life Cycle Assesment). solutions database connected to external databases like the one provided by the ITeC [42].

The main output variables are shown in Table2 and represent only a few subsets of the information obtained from the model that is stored in a text file following an XML format. Sustainability 2017, 9, 1004 7 of 17

Table 2. Main output variables.

Variable Description The energy demand of the model will be determined to minimize the Energy Impact Sustainability 2017, 9, 1004 building’s energy (energy for heating and cooling). 7 of 17 Analysis of the environmental impact of materials used (Global Warming, Environmental ImpactAnalysis of the economic costs of the LCA process, the materials used in construction, Ozone Depletion, etc.) according to an LCA (Life Cycle Assesment). Economic Impact and the energy and material demand of the prototype (described in the standard prEN 15643-4 [43]Analysis). of the economic costs of the LCA process, the materials used in Economic ImpactAnalysisconstruction, of the social impacts and the of energy the building and material on its immediate demand of environment the prototype (impacts Social Impact (described in the standard prEN 15643-4 [43]). can be positive or negative) Analysis of the social impacts of the building on its immediate environment Social Impact An example of the SDL(impacts PROCESS can that be positive rules the or negative)model behavior (Figure 4) shows the behavior for the DESIGNED state of the P1_Building PROCESS. This is the process responsible for the representationAn example ofof thethe SDLentire PROCESS life of the building that rules, from the the model design behavior phase to (Figure the destruction4) shows and the behaviorrecycling for the DESIGNEDphases. In thisstate case, of the buildingP1_Building startsPROCESS. in the DESIGNED This is the state, process representing responsible the forfactthe that representation a specific of thedesign entire has life been of the selected building, for from the building the design and phase the building to the destruction process can and begin. recycling The BUILD phases. state In this represents the fact that the building has been constructed and can be used. The USED state represents case, the building starts in the DESIGNED state, representing the fact that a specific design has been the fact that the useful life of the building is over and it needs to be demolished. Finally, the selected for the building and the building process can begin. The BUILD state represents the fact that DESTROYED state represents the fact that the building has been demolished. Every one of these the building has been constructed and can be used. The USED state represents the fact that the useful states has its own calculations and logistics; some of them using external calculus engines in a co- life ofsimulation the building scenario. is over and it needs to be demolished. Finally, the DESTROYED state represents the fact thatThe thestandard building ISO has14040 been was demolished. adopted for Every the LCA one and of these impact states method has itsanalysis. own calculations Specifically, and logistics;impact some method of them ECO using99 [44]external was used. calculus engines in a co-simulation scenario.

FigureFigure 4. 4Behavior. Behavior ofof thethe building in in the the DESIGNED DESIGNED state. state.

Auxiliary variables defined in DCL allowed us to represent the parametrizations for a house in an experiment, which helps in defining the experimental design. The main model elements are represented in PROCESS diagrams; PROCEDURES are used to calculate the building energy Sustainability 2017, 9, 1004 8 of 17

The standard ISO 14040 was adopted for the LCA and impact method analysis. Specifically, impact method ECO 99 [44] was used. Auxiliary variables defined in DCL allowed us to represent the parametrizations for a house Sustainabilityin an experiment, 2017, 9, 1004 which helps in defining the experimental design. The main model elements8 of are 17 represented in PROCESS diagrams; PROCEDURES are used to calculate the building energy consumed consumed(this part of (this the part model of the can model analyze can and analyze optimize and differentoptimize HVACdifferent systems HVAC based systems on thebased COP on andthe COPthe performance and the performance of each active of eachclimate active machine climate where machine we know where the we kWh/m know2 energythe kWh/m consumption2 energy consumptionrequired to reach required the thermal to reach comfort the thermal for each comfort situation for studied). each situation Once thestudied). model O isnce correct, the model the final is correct,users can the only final modify users can the DCL’sonly modify of the modelthe DCL’s by using of the a model cloud infrastructure.by using a cloud infrastructure. We can analyze the behavior of the PROCEDURES ( (thatthat use a database to calculate the environmentalenvironmental,, social social,, and and economic economic aspects aspects of ofthe the materials materials used used),), and andthe systems the systems applied applied to build to thebuild model the model.. The TheS1_ACVS1_ACV-Materials-Materials PROCEDUREPROCEDURE analyzes analyzes the the environmental environmental impacts impacts and and the economic and social costs related to materials. The The S1_ConstructionProcessS1_ConstructionProcess PROCEDURE analyzes the construction process, as seenseen in Figure5 5..

Figure 5 5.. ThisThis figure figure details the S1_ACV_Materials and S1_ConstructionProcess.

One can connect,connect, thanksthanks to to the the SDL SDL modularity, modularity, several several buildings buildings to defineto define an urbanan urban area. area. In this In thiscase, case, the modelthe model only only needs needs to define to define the communication the communication mechanisms mechanisms to be usedto be (onused the (on model the model itself). itself).The execution The execution of the modelof the model that contains that contains several several buildings buildings provides provides information information of an of urban an urban area, area,instead instead of providing of providing information information of just of a just single a single isolated isolated building. building. CoCo-simulation-simulation techniques techniques can can be be applied applied on on PROCEDURES PROCEDURES to calculate to calculate the energy the energy impacts impacts using usingstate-of-the-art state-of-the calculus-art calculus engines engines like Energy+like Energy+ [45 ].[45] Heuristic. Heuristic optimization optimization is is used, used, based based onon Hill Climbing, Simulated AnnealingAnnealing,, or NSGAII algorithms. Th Theseese optimization techniques allow the dramatic reduc reductiontion of the number of of scenarios scenarios to to be be executed, executed, whilst whilst still still obtaining obtaining a a good good answer. answer. The timetime neededneeded toto obtain obtain an an answer answer for for an an experimental experimental design design can can be large,be large as suggested, as suggested in [46 in] who[46] whopresented presented an example an example of an of execution an execution of a real of a experiment real experiment in a distributed in a distributed scenario. scenario. The criteria The criteria used wasused based was based on an expressionon an expression that combined that combine the differentd the different variables variables that were that interesting were interesting in the definition in the definitionof the LCA. of This the expressionLCA. This wasexpression defined was by the defi userned in by the the software user in we the used software (see next we section),used (see which next section)defined, thewhich experimental defined the design experimental and how design it was and executed. how it was executed.

4. Results The model was was implemented implemented using using SDLPS SDLPS software software [47,48] [47,48.]. SDLPS SDLPS was was built built using using C++ C++ and and C languages.C languages. The The model model code code (written (written in C in for C the for tasks the tasks and procedures and procedures of the of SDL the blocks) SDL blocks) was used was throughused through a Dynamic a Dynamic Link Link Library Library (DLL). (DLL). The The model model was was defined defined directly directly using using Microsoft Microsoft Visio®,, following the standard imimplementedplemented on the 2013 version as SDLPS understands this format and transforms it it to an XML format that represents the SDL model. This coding implies that the model can be validated and verifiedverified by reviewing the graphic diagrams inin Microsoft Visio®® . This This pr propertyoperty dramatically simplifies simplifies the interaction between the different actors involved in the project. Figure 66 shows the SDLPS loaded with a model for a building. Sustainability 2017, 9, 1004 9 of 17 Sustainability 2017, 9, 1004 9 of 17 Sustainability 2017, 9, 1004 9 of 17

Figure 6. SDLPS interface with the building model. FigureFigure 6.6.SDLPS SDLPS interfaceinterface withwith thethe buildingbuilding model.model. 5. Executing on the Cloud: NECADA 5.5. ExecutingExecuting onon thethe Cloud:Cloud: NECADANECADA The model was defined using SDL and all team members were aware of the hypotheses used in it. However,TheThe modelmodel in thiswaswas type defineddefined of model using ,SDL parametrization and and all all team team membersis members a key element were were aware aware that ofneeds of the the hypothesesto hypotheses be reviewed used used to in incarefullyit. it.However, However, assure in in thatthis this thetype type data of of modelintroduced model,, parametrization parametrization are accurate andis is a a correct.key key element element Additionally that that needs, once toto bethebe reviewed revieweddeveloping toto carefullyteamcarefully has assureagreeassured thatonthat the thethe validity datadata introducedintroduced of the model areare accurate(inaccurate some andsenseand correct.correct. the model Additionally,Additionally is accredited),once once, nothe the modification developingdeveloping teamisteam needed has has agreed agreeuntil dthe on on underlying thethe validityvalidity hypotheses ofof thethe modelmodel are (in(in modified. somesome sensesense In thethatthe modelmodel case, the isis accredited),accredited) parametrization, nono modificationmodification proposed isonis needed neededthe SDL untiluntil diagrams thethe underlying underlying trough the hypotheseshypotheses DCL’s is aretheare modified.keymodified. element InIn thatinthat defin case,caseing, the the the parametrization parametrization different scenarios proposedproposed to be onconducted.on thethe SDLSDL This diagramsdiagrams parametrization troughtrough thethe DCL’scanDCL’s be isdoneis thethe by keykey third elementelement part inyin teams definingdefin ingwho thethe do differentdifferent not have scenariosscenarios to be directly toto bebe conducted.connectedconducted. with ThisThis the parametrizationparametrization model development. can be donedone byby thirdthird partyparty teamsteams whowho dodo notnot havehave toto bebe directlydirectly connectedconnected withwith thethe modelmodel development.development. 5.1. Description of the Current Platform 5.1.5.1. DescriptionDescription ofof thethe CurrentCurrent PlatformPlatform To simplify parametrization and error reduction, a Cloud service named NECADA® was ® developedToTo simplify simplify to allow parametrization parametrization model parametrization and and error error and reduction, reduction execution., aa CloudFigure Cloud service7 service shows named namedthe login NECADA NECADA screen of® was wasthe developedNECADAdeveloped cloud toto allowallow solution. modelmodel parametrizationparametrization andand execution.execution. FigureFigure7 7 shows shows the the login login screen screen of of the the NECADANECADA cloudcloud solution.solution.

Figure 7. NECADA cloud solution login screen. Figure 7. NECADA cloud solution login screen. Figure 7. NECADA cloud solution login screen. Sustainability 2017, 9, 1004 10 of 17

NECADA manages the parametrization of the experimental design to be executed and sends it to SDLPS, which executes the LCA building model. SDLPS acts as a simulation server executing the experiments defined over the web; Figure 8 shows the public projects of the NECADA website. Each one of these projects owns specific building models (the building geometry), based on the SustainabilityBIM methodology2017, 9, 1004 [49], materials, constructing solution, weathers, and orientations that can be10 of 17 permutated on the experiments. These permutations are the key elements that are defined by the analyst since it defines the space to be explored by the model (Figure 9). NECADAWith this manages information the parametrization—and based on the of SDL the experimentalmodel that defines design the toLCA be executedof a building and— sendsthe it to SDLPS,LCA could which be executescalculated the for LCA the different building scenar model.ios. SDLPS This allow actsed as us a simulationto obtain the server Pareto executing frontier, the experimentsdepending defined on the overselected the variables web; Figure (see8 Figureshows 10 the to public review projects the file that of the contain NECADAed the website.indicators usedEach in one the ofanalysis these). projectsThese indicators owns specific were defined building formally models on the (the SDL building model and geometry), filled with based the on the BIMcorrect methodology values using [the49], database materials, the modeler constructing used on solution, the specific weathers, project. and orientations that can be permutatedIn Figure on the 11 experiments., the results obtained These from permutations NECADA arefor the e(CO) key elements building thatare shown are defined [50]. The by the analystPareto since frontier it defines represented the space in the to beupper explored left area by of the the model figure, (Figure allowed9). us to determine the best fit among all the possible alternatives in the model.

FigureFigure 8. Public 8. Public projects projects shown shown on on thethe NECADANECADA website website include include the the BRGF, BRGF, MARIE MARIE,, and and (e)CO (e)CO projects.projects By.using By using NECADA, NECADA the, the users users do notdo not interact interact directly directly with with the the SDL SDL model, model but, but with with the the factors (declarations)factors (declarations) used inside used the inside model the to model parametrize to parametrize the different the different experiments. experiments. Sustainability 2017, 9, 1004 11 of 17

FigureFigure 9. File 9. File containing containing the the permutations permutations toto bebe executed and and those those defined defined in the in theexperimental experimental design. design.

Figure 10. Indicators to be used on the calculus of the LCA; not all the indicators are used in this specific example. Sustainability 2017, 9, 1004 11 of 17

Sustainability 2017, 9, 1004 11 of 17

With this information—and based on the SDL model that defines the LCA of a building—the LCA could be calculated for the different scenarios. This allowed us to obtain the Pareto frontier, depending on the selected variables (see Figure 10 to review the file that contained the indicators used in the analysis). These indicators were defined formally on the SDL model and filled with the correct values usingFigure the database9. File containing the modeler the permutations used on the to specificbe executed project. and those defined in the experimental design.

FigureFigure 10.10. Indicators to be used on the calculuscalculus of thethe LCA;LCA; not allall thethe indicatorsindicators areare usedused inin thisthis specificspecific example.example.

In Figure 11, the results obtained from NECADA for the e(CO) building are shown [50]. The Pareto frontier represented in the upper left area of the figure, allowed us to determine the best fit among all Sustainability 2017, 9, 1004 12 of 17 the possible alternatives in the model.

Figure 11. Results from the analysis of the different scenario for the e(CO) building. On top are shown the Pareto frontiers for the more than 4000 different parametrizations analyzed, on the bottom is a sensitivityFigure 11 analysis. Results obtainedfrom the analysis from a detailed of the different analysis scenario of all the for obtained the e(CO) information. building. On Thistop are helps shown in the the Pareto frontiers for the more than 4000 different parametrizations analyzed, on the bottom is a determination of optimal actions to save energy using the passive elements in the building. sensitivity analysis obtained from a detailed analysis of all the obtained information. This helps in the determination of optimal actions to save energy using the passive elements in the building.

Detailed results obtained on the (e)CO project can be found in [40]. Thanks to this holistic solution, the e(CO) building LCA can be improved in its design phase. NECADA simplified the interaction of the non-specialists in the simulation user-group with an SDL model that only needed to parametrize the different factors that must be introduced in each experiment. Through the use of SDL, we could go further in our analysis. During the use of the building, we could take advantage of the different sensors the houses currently had, from the real-time analysis of energy consumption to the possibility of controlling the windows. Since a model that defines the optimum solution for an LCA is mainly focused on the planning phase, some deviations and corrections must be performed during the use phase of the building. Here, a Smart Home connected through SDL can achieve its full potential.

5.2. A Smart House Connected through SDL and IoT In the “use” phase (Figure 3), we focused on how people were using the houses and how energy was managed. A Smart Home is usually seen as a house where almost everything can be controlled, including the windows, doors, temperature, etc. The concept of smart home was connected through the Cradle to Cradle metaphor to the sustainability concept. The interconnection of different hardware and software used to control a house makes it difficult to easily extend and reuse the solutions. Plenty of different alternatives currently exist, some of them based on standards like KNX [51,52], standardized on EN 50090, ISO/IEC 14543, or BACnet [53], an ASHRAE, ANSI, and ISO 16484-5 standard protocol, as well as others used as de facto standards like Modbus, a serial communications protocol originally published by Modicon (now Schneider Electric) in 1979 for use with its programmable logic controllers (PLCs). There are also many other alternatives like LonWorks, Home automation, BACnet, DOLLx8, EnOcean INSTEON, Z-Wave, Intelligent building, Lighting control system, OpenTherm, Room automation, Smart Environments, and Touch panel. In this scenario, with several communication protocols interacting with several devices, the use of an abstract layer that helps in the definition and formalization of the several processes that must be considered in a smart home, or as an extension of this in a smart city, is necessary. Additionally, as shown in this paper, SDL can be used to represent the LCA of a building, becoming an interesting Sustainability 2017, 9, 1004 12 of 17

Detailed results obtained on the (e)CO project can be found in [40]. Thanks to this holistic solution, the e(CO) building LCA can be improved in its design phase. NECADA simplified the interaction of the non-specialists in the simulation user-group with an SDL model that only needed to parametrize the different factors that must be introduced in each experiment. Through the use of SDL, we could go further in our analysis. During the use of the building, we could take advantage of the different sensors the houses currently had, from the real-time analysis of energy consumption to the possibility of controlling the windows. Since a model that defines the optimum solution for an LCA is mainly focused on the planning phase, some deviations and corrections must be performed during the use phase of the building. Here, a Smart Home connected through SDL can achieve its full potential.

5.2. A Smart House Connected through SDL and IoT In the “use” phase (Figure3), we focused on how people were using the houses and how energy was managed. A Smart Home is usually seen as a house where almost everything can be controlled, including the windows, doors, temperature, etc. The concept of smart home was connected through the Cradle to Cradle metaphor to the sustainability concept. The interconnection of different hardware and software used to control a house makes it difficult to easily extend and reuse the solutions. Plenty of different alternatives currently exist, some of them based on standards like KNX [51,52], standardized on EN 50090, ISO/IEC 14543, or BACnet [53], an ASHRAE, ANSI, and ISO 16484-5 standard protocol, as well as others used as de facto standards like Modbus, a serial communications protocol originally published by Modicon (now Schneider Electric) in 1979 for use with its programmable logic controllers (PLCs). There are also many other alternatives like LonWorks, Home automation, BACnet, DOLLx8, EnOcean INSTEON, Z-Wave, Intelligent building, Lighting control system, OpenTherm, Room automation, Smart Environments, and Touch panel. In this scenario, with several communication protocols interacting with several devices, the use of an abstract layer that helps in the definition and formalization of the several processes that must be considered in a smart home, or as an extension of this in a smart city, is necessary. Additionally, as shown in this paper, SDL can be used to represent the LCA of a building, becoming an interesting candidate to represent all the processes related to a building life from the planning phase to the day to day use of the different devices interacting in a home; therefore, giving information to the residents regarding the deviations detected from the original planning. For a detailed description of the proposed use of SDL in the frame of IoT, [54] can be consulted.

6. Discussion The proposed methodology has been successfully used in several projects. To design a double active façade [55], we used the model (in combination with Computational Fluid Dynamics (CFD) models) in a co-simulation approach. In the European project MARIE [56–58], we used the model to perform a double analysis for the residential typologies in the Mediterranean area, first by optimizing the comfort and economic criteria based on passive measures and, second, by a cost-effective analysis selecting active energy efficiency measures. Other projects where this model has been used include the ACE project described in [59], where the model was used to simulate the behavior of the different typologies that were later implemented in the project. The model, using different calculus engines, allowed the optimization of the analyzed system following the European normative CEN/TC 350 (UNE-EN 15643-2, UNE-EN 15643-3, UNE-EN 15643-4). The co-simulation approach made it possible to change the calculus engine to be used, for example, in the MARIE project, we used Trnsys® [60] while in others we used Energy+ [45]. Hence, the model could be used for normative purposes and to obtain green certification for new or refurbished buildings. The proposed methodology allowed us to define the model using a holistic approach and simplify the interaction between the different specialists. By using an infrastructure that understood this formal Sustainability 2017, 9, 1004 13 of 17

candidate to represent all the processes related to a building life from the planning phase to the day to day use of the different devices interacting in a home; therefore, giving information to the residents regarding the deviations detected from the original planning. For a detailed description of the proposed use of SDL in the frame of IoT, [54] can be consulted.

6. Discussion The proposed methodology has been successfully used in several projects. To design a double active façade [55], we used the model (in combination with Computational Fluid Dynamics (CFD) models) in a co-simulation approach. In the European project MARIE [56–58], we used the model to perform a double analysis for the residential typologies in the Mediterranean area, first by optimizing the comfort and economic criteria based on passive measures and, second, by a cost-effective analysis selecting active energy efficiency measures. Other projects where this model has been used include the ACE project described in [59], where the model was used to simulate the behavior of the different typologies that were later implemented in the project. The model, using different calculus engines, allowed the optimization of the analyzed system following the European normative CEN/TC 350 (UNE-EN 15643-2, UNE-EN 15643-3, UNE-EN 15643- 4). The co-simulation approach made it possible to change the calculus engine to be used, for example, in the MARIE project, we used Trnsys® [60] while in others we used Energy+ [45]. Hence, the model could be used for normative purposes and to obtain green certification for new or refurbished buildings. Sustainability 2017, 9, 1004 13 of 17 The proposed methodology allowed us to define the model using a holistic approach and simplify the interaction between the different specialists. By using an infrastructure that understood language,this formal no language, specific implementationno specific implementation was performed was onperformed the projects, on the thus projects, simplifying thus simplifying the verification the processverification required process for therequired simulation for the projects. simulation The projects. steps depicted The steps in red depicted in the simplified in red in the version simplified of the modelingversion of processthe model presenteding process by [ 41presented], are those by that[41], areare improved,those that seeare Figureimproved, 12. see Figure 12.

Figure 12.12. SimplifiedSimplified version of the modelingmodeling process, basedbased onon [[41]41].. TheThe SDLSDL executableexecutable modelmodel cancan be validated validated by by all team all team members; members; furthermore, furthermore, the automatic the automatic implementation implementation assures its verifiability. assures its verifiability. 7. Conclusions 7. Conclusions In this paper, the use of SDL to represent the LCA of a building holistically was introduced. We presentedIn this paper, how the this use methodology of SDL to represent simplified the the LCA validation of a building and verificationholistically was problems introduced. related We to thispresented type of how complex this methodology model. It was simplified also shown the validation that the implementation and verification of problems these models related could to this be achievedtype of complex usinga model. desktop It was program also shown that implemented that the implementation SDL models of (in these our models case SDLPS) could andbe achieved a cloud infrastructureusing a desktop that p usedrogram SDLPS that as implemented a simulation server SDL models to offer a(in cloud our solution. case SDLPS This)cloud and solution a cloud simplifiedinfrastructure the that parametrization used SDLPS processas a simulation of a building. serverThe to offer use ofa cloud a standard solution. language, This cloud such solution as SDL, simplifiedsimplified implementation;the parametrization as severalprocess other of a building. tools understand The use SDL,of a standard as mentioned language, in [61 su–63ch], as among SDL, many others, this made the definition of the model and the final implementation independent of the chosen infrastructure. Model validation was performed with the SDL representation of the model, so that all project members involved could participate in the validation process. Verification was assured as the tool recognizes SDL diagrams; however, other validation techniques could also be applied (black box, white box, Turing test, etc.). From the perspective of the language proposed (one can select to follow this methodology with other formal languages like DEVS or PetriNets among others), SDL could be used not only to model the LCA of a building, but could naturally also represent the existing communications between the different sensors available in a building [54]. This can complete the vision of the building allowing models to be used not only at design phase, but also in the operative phase, thus becoming a virtual advisor for a house or urban area.

Author Contributions: Pau Fonseca i Casas and Antoni Fonseca i Casas conceived the use of SDL in this study and adapted, designed, and implemented the methodology and the tools described here. The work was based on a simulator named SDLPS, developed initially by Pau in his Ph.D., that executed the model developed in Antoni’s Ph.D. Conflicts of Interest: The authors declare no conflict of interest. Sustainability 2017, 9, 1004 14 of 17

Abbreviations The following abbreviations are used in this manuscript:

ANSI American National Standards Institute ASHRAE American Society of Heating, Refrigerating, and Air-Conditioning Engineer BRGF Biblioteca Sector Gabriel Ferraté (e)CO Equilibrium through Cooperation ISO International Organization for Standardization LCA Life Cycle Assesment SDL Specification and Description Language

References

1. IEA. Energy Efficiency Market Report; IEA: Paris, France, 2014. 2. ICER. Energy Efficiency Potential in the U.S.: 2012 through 2035. Available online: http://www.icer- regulators.net/portal/page/portal/ICER_HOME/publications_press/ICER_Chronicle/Art2_10a.html (accessed on 2 May 2017). 3. Van der Grintern, B. Cradle to Cradle in a Nutshell. 2008. Available online: http://www.bluehaired.com/ corner/wp-content/uploads/2009/12/Cradle-to-Cradle-in-a-nutshell-Bram-van-der-Grinten-2008.pdf (accessed on 2 May 2017). 4. Rønning, A.; Brekke, A. Life cycle assessment (LCA) of the building sector: Strengths and weaknesses. In Eco-Efficient Construction and Building Materials; Woodhead Publishing: Sawston, UK, 2014; pp. 63–83. 5. Hollberg, A.; Ruth, J. LCA in architectural design—A parametric approach. Int. J. Life Cycle Assess. 2016, 21, 943–960. [CrossRef] 6. Means, P.; Guggemos, A. Framework for Life Cycle Assessment (LCA) Based Environmental Decision Making During the Conceptual Design Phase for Commercial Buildings. Proced. Eng. 2015, 118, 802–812. [CrossRef] 7. Vilches, A.; Garcia-Martinez, A.; Sanchez-Montañes, B. Life cycle assessment (LCA) of building refurbishment: A literature review. Energy Build. 2016, 135, 286–301. [CrossRef] 8. Anand, C.; Amor, B. Recent developments, future challenges and new research directions in LCA of buildings: A critical review. Renew. Sustain. Energy Rev. 2017, 67, 408–416. [CrossRef] 9. BPIE. Principles for Nearly Zero-Energy Buildings. Paving the Way for Effective Implementation of Policy Requirements; Buildings Performance Institute Europe: Brussels, Belguim, 2011. 10. Salom, J.; Widén, J.; Candanedo, J.; Sartori, I.; Voss, K.; Marszal, A. Understanding net zero energy buildings: Evaluation of load matching and grid interaction indicators. In Proceedings of the Building Simulation 2011: 12th Conference of International Building Performance Simulation Association, Sydney, Australia, 14–16 November 2011. 11. Sartori, I.; Napolitano, A.; Voss, K. Net zero energy buildings: A consistent definition framework. Energy Build. 2012, 48, 220–232. [CrossRef] 12. Salom, J.; Cubí, E.; Sánchez, A. Busando una definición concreta para un objetivo amplio. CIC Arquitectura y Construcción 2011, 485, 54–57. 13. Salom, J.; Cubí, E.; Sartori, I. Edificio de energía cero: Definiciones e interacción con las redes energéticas. In Libro de Comunicaciones I Congreso EECN; EECN: Madrid, Spain, 2012. 14. Llera-Sastresa, E.; Scarpellini, S.; Rivera-Torres, P.; Aranda, J.; Zabalza-Bribián, I.; Aranda-Usón, A. Energy Vulnerability Composite Index in Social Housing, from a Household Energy Poverty Perspective. Sustainability 2017, 9, 691. [CrossRef] 15. EeBGuide. The EeBGuide Project. Operational Guidance for Life Cycle Assessment Studies of the Energy Efficient Buildings Initiative; Fraunhofer-Gesellschaft: Munich, Germany, 2013. 16. SUDOE. EnerBuiLCA. 2012. Available online: http://4.interreg-sudoe.eu/contenido-dinamico/libreria- ficheros/8656A5AF-B808-1CEB-5C83-37836BB24FAA.pdf (accessed on 7 June 2017). 17. CEN/TC 350. Sustainability of Construction Works; European Committee for Standardization: Brussels, Beguim, 2012. Sustainability 2017, 9, 1004 15 of 17

18. Laëtitia, A.; Olivier, B.; Pascal, R.; Daniel, Q.; Pascal, R. A simple method to consider energy balance in the architectural design of residential buildings. In Proceedings of the Symposium on Simulation for Architecture and Urban Design, Boston, MA, USA, 3–7 April 2011; pp. 115–122. 19. Sattrup, P.A.; Strømann-Andersen, J.B. A methodological study of environmental simulation in architecture and engineering: Integrating daylight and thermal performance across the urban and building scales. In Proceedings of the 2011 Symposium on Simulation for Architecture and Urban Design, Boston, MA, USA, 3–7 April 2011; Attar, R., Ed.; Society for Computer Simulation International: San Diego, CA, USA, 2011; pp. 115–123. 20. Sodja, A.; Zupanˇciˇc,B. Modelling thermal processes in buildings using an object-oriented approach and Modelica. Simul. Model. Pract. Theory 2009, 17, 1143–1159. [CrossRef] 21. Counsell, J.M.; Khalid, Y.A.; Brindley, J. Controllability of buildings: A multi-input multi-output stability assessment method for buildings with slow acting heating systems. Simul. Model. Pract. Theory 2011, 19, 1185–1200. [CrossRef] 22. Liu, B.; Duan, S. Energy efficiency evaluation of building integrated photovoltaic systems with different power configurations. Simul. Model. Pract. Theory 2012, 29, 93–108. [CrossRef] 23. Bruno, M.; Henderson, K.; Kim, H. Multi-objective optimization in urban design. In Proceedings of the 2011 Symposium on Simulation for Architecture and Urban Design, Boston, MA, USA, 3–7 April 2011; Attar, R., Ed.; Society for Computer Simulation International: San Diego, CA, USA, 2011; pp. 102–109. 24. Zeigler, B.P.; Praehofer, H.; Kim, T.G. Theory of Modeling and Simulation Handbook of Simulator-Based Training Creating Computer Simulation Systems: An Introduction to the High Level Architecture; Academic Press: Waltham, MA, USA, 2000. 25. ITU-T. Specification and Description Language—Overview of SDL-2010; ITU: Geneva, Switzerland, 2011. 26. Doldi, L. SDL Illustrated—Visually Design Executable Models; TMSO Multimedia: Toulouse, France, 2001. 27. Cabasino, M.P.; Giua, A.; Seatzu, C. Introduction to Petri Nets. Lect. Notes Control Inf. Sci. 2013, 433, 191–211. 28. Ahmed, A.S.; Wainer, G.; Mahmoud, S. Integrating building information modeling & cell-DEVS simulation. In Proceedings of the 2010 Spring Simululation Multiconference, Orlando, FL, USA, 11–15 April 2010; Society for Computer Simulation International: San Diego, CA, USA, 2001. [CrossRef] 29. Trˇcka,M.; Hensen, J.L.M.; Wetter, M. Co-simulation for performance prediction of integrated building and HVAC systems—An analysis of solution characteristics using a two-body system. Simul. Model. Pract. Theory 2010, 18, 957–970. [CrossRef] 30. Stahl, I.; Born, R.G.; Henriksen, J.O.; Herper, H. GPSS 50 years old, but still young. In Proceedings of the 2011 Winter Simulation Conference, Phoenix, AZ, USA, 11–14 December 2011; pp. 3947–3957. 31. O’Reilly, J.J. Optimization test problems with uniformly distributed coefficients. In Proceedings of the 1991 Winter Simulation Conference, Phoenix, AZ, USA, 8–11 December 1991. 32. Jones, J. SIMPROCESS III: Object-oriented business process simulation. In Proceedings of the 1995 Winter Simulation Conference, Arlington, VA, USA, 3–6 December 1995. 33. Altiok, T.; Melamed, B. Simulation Modeling and Analysis with ARENA; Academic Press: Waltham, MA, USA, 2007. 34. Pegden, C.D.; Sturrock, D.T. Introduction to Simio. In Proceedings of the 2011 Winter Simulation Conference, Phoenix, AZ, USA, 11–14 December 2011; pp. 29–38. 35. Concepcion, A.I.; Zeigler, B.P. DEVS formalism: A framework for hierarchical model development. IEEE Trans. Softw. Eng. 1988, 14, 228–241. [CrossRef] 36. Vangheluwe, H.L.M. DEVS as a common denominator for multi-formalism hybrid systems modelling. In Proceedings of the IEEE International Symposium on Computer-Aided Control System Design (Cat. No.00TH8537), Anchorage, AK, USA, 25–27 September 2000; IEEE: New York, NY, USA, 2000; pp. 129–134. 37. Shang, H.; Wainer, G. A Simulation Algorithm for Dynamic Structure DEVS Modeling. In Proceedings of the 2006 Winter Simulation Conference, Monterey, CA, USA, 3–6 December 2006; pp. 815–822. 38. Goldstein, R.; Khan, A.; East, K.S. Development of discrete event system specification (DEVS) building performance models for building energy design. In Proceedings of the Symposium on Simulation for Architecture & Urban Design, San Diego, CA, USA, 7–10 April 2013. 39. Gotzhein, R.; Krämer, M.; Litz, L.; Chamaken, A. Energy-aware system design with SDL. In Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Springer: Berlin, Germany, 2009; pp. 19–33. Sustainability 2017, 9, 1004 16 of 17

40. Fonseca i Casas, P.; Fonseca i Casas, A.; Garrido-Soriano, N.; Casanovas, J. Formal simulation model to optimize building sustainability. Adv. Eng. Softw. 2014, 69, 62–74. [CrossRef] 41. Sargent, R. Verification and validation of simulation models. In Proceedings of the 2009 Winter Simulation Conference, Austin, TX, USA, 13–16 December 2009; Rossetti, M.D., Hill, R.R., Johansson, B., Dunkin, A., Ingalls, R.G., Eds.; IEEE: New York, NY, USA, 2009; pp. 162–176. 42. Alfaro Garrido, L.; Lucas Masero, J.; Díez Bernabé, G. New tool to identify environmental impacts on the construction. In Proceedings of the VII Congreso Europeo sobre Eficiencia Energética y Sostenibilidad en Arquitectura y Urbanismo, San Sebastian, Spain, 4–6 July 2016. 43. AEN/CTN. Evaluación de la Sostenibilidad en los Edificios. Parte 4: Marco para la Evaluación del Comportamiento Económico; UNE-EN 15643-4:2012; AENOR: Madrid, Spain, 2012. 44. Baayen, H. Eco-indicator 99 Manual for Designers; Ministry of Housing, Spatial Planning and the Environment: The Hague, The Netherlands, 2000. 45. EnergyPlus. Input Output Reference the Encyclopedic Reference to EnergyPlus Input and Output; U.S. Department of Energy: Washington, DC, USA, 2014. 46. Fonseca i Casas, P.; Fonseca i Casas, A.; Garrido-Soriano, N.; Ortiz, J.; Casanovas, J.; Salom, J. Optimal Buildings’ Energy Consumption Calculus through a Distributed Experiment Execution. Math. Probl. Eng. 2015, 2015, 1–12. [CrossRef] 47. Fonseca i Casas, P. SDL distributed simulator. In Proceedings of the 2008 Winter Simulation Conference, Miami, FL, USA, 7–10 December 2008. 48. Fonseca i Casas, P.; Pi Palomés, X.; Casanovas Garcia, J.; Jové, J. Definition of Virtual Reality Simulation Models Using Specification and Description Language Diagrams; Khendek, F., Toeroe, M., Gherbi, A., Reed, R., Eds.; Lecture Notes in Computer Science; Springer: Berlin, Germany, 2013. 49. Smith, M. BIM in Construction. Available online: http://www.thenbs.com/topics/BIM/articles/ bimInConstruction.asp (accessed on 7 June 2017). 50. Fonseca i Casas, A.; Fonseca i Casas, P.; Colls, M. In NZEB. Optimización mediante co-simulación. Estudio de caso (e)Co, proyecto prototipo del concurso Solar Decathlon 2012. In Proceedings of the CONAMA 2012 Congreso Nacional de Medio Ambiente, Madrid, Spain, 26–30 November 2012. 51. KNX Association. KNX System Specifications—Architecture; KNX Association: Brussels, Belguim, 2013. 52. KNX Association. System Specifications; KNX Association: Brussels, Belguim, 2014. 53. ISO. Building Automation and Control Systems (BACS)—Part 1: Project Specification And Implementation; ISO: Geneva, Switzerland, 2010. 54. Sherratt, E.; Ober, I.; Gaudin, E.; Fonseca i Casas, P.; Kristoffersen, F. SDL—The IoT language. In SDL 2015: Model-Driven Engineering for Smart Cities. Lecture Notes in Computer Science; Fischer, J., Scheidgen, M., Schieferdecker, I., Reed, R., Eds.; Springer: Cham, Switzerland, 2015. 55. ACCIONA. Japan-Spain Energy Efficient Development for Ultra-Low Energy Buildings. Available online: https://www.acciona.com/sustainability/environment/environmental-innovation/systems-storing- seasonal-heat-buildings/ (accessed on 7 June 2017). 56. Salom, J.; Ortiz, J. Russo, Method to develop cost-effective studies of energy efficiency measures for Mediterranean residential existing buildings with multi-criteria optimization. In Proceedings of the World Sustainable Buldings SB14, Barcelona, Spain, 28–30 October 2014. 57. Ortiz, J.; Fonseca i Casas, A.; Salom, J.; Garrido Soriano, N.; Fonseca i Casas, P. Cost-effective analysis for selecting energy efficiency measures for refurbishment of residential buildings in Catalonia. Energy Build. 2016, 128, 442–457. [CrossRef] 58. Ortiz, J.; Fonseca, A.; Salom, J.; Garrido, N.; Fonseca, P.; Russo, V. Comfort and economic criteria for selecting passive measures for the energy refurbishment of residential buildings in Catalonia. Energy Build. 2016, 110, 195–210. [CrossRef] 59. Fonseca i Casas, A.; Fonseca i Casas, P. Casanovas, Analysis of Applications to Improve the Energy Savings in Residential Buildings Based on Systemic Quality Model. Sustainability 2016, 8, 1051–1069. [CrossRef] 60. Thermal Energy System Specialists. TRNSYS Transient System Simulation Tool. Available online: http: //www.trnsys.com/ (accessed on 21 March 2015). 61. PragmaDev SARL. PragmaDev Studio. Available online: http://www.pragmadev.com/product/index.html (accessed on 9 January 2016). Sustainability 2017, 9, 1004 17 of 17

62. Cinderella ApS. Cinderella. Available online: http://www.cinderella.dk/ (accessed on 9 January 2016). 63. IBM Co. Rational SDL Suite. Available online: http://www-03.ibm.com/software/products/en/ratisdlsuit (accessed on 9 January 2016).

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).