Final Report Summary - WEB2SME (Web 2.0 driven service-builder for product extension in globally-acting SMEs)
Executive summary:
The objective of the project is to provide an innovative comprehensive software (SW) solution, following the Web 2.0 principles, to extend products of automation systems and electrical equipment manufacturer small and medium-sized enterprises (SMEs) acting on the global market.
WEB2SME solution will allow:
(1) industrial user SME - first users of such SW system to strengthen their competitiveness at the global market by providing new product and customer support services and enabling new business models;
(2) information and communication technology (ICT) and service engineering vendor SMEs to provide a new SW system highly competitive at the ICT market.
WEB2SME consortium includes 2 research institutes, 3 ICT vendors and 3 user SMEs, equipment and automation system providers located in three European countries namely Spain, Poland and Germany.
The concept behind WEB2SME is based on the need on user SMEs to extend their products with associated services product extension services (PES) with different product and customer support services, such as condition based maintenance, problem solving, equipment reconfiguration services etc., and to be able to cost effectively provide these services to customers distributed worldwide.
An essential aspect of such PES is that they require a means for efficient collaboration of different actors in a supply chain and customers, within a geographically distributed extended enterprise context. Users need ICT solutions which will allow smooth transition from the current (mostly locally oriented) service provision, to provision of services at global market, partly in 'virtual world'.
The WEB2SME methodology shall serve as guideline for SMEs, which intend to apply the WEB2SME approach - combining collaborative services and KM technology to extend products in SMEs - addressing both technological and organisational aspects related to product extensions and knowledge sharing within the EE in the global market.
Within the project it is planned to realise a comprehensive package of methods, guidelines and manuals which can be divided into the following parts:
(a) methodology to set up and maintain a collaborative working environment (CWE);
(b) methodology to set up new and maintain existing PES; and
(c) manuals.
Within the ICT system concept, the different main components, namely the core collaborative services (CCS) and services / tools for knowledge sharing, will be implemented as well as the concept of PES and how they can be realised based on a combination of core services and different application specific tools.
WEB2SME consortium includes 3 ICT vendor SMEs (INNOPOLE, ESCO, freiheit.com) whose business is to provide novel ICT solutions and support SME users to develop / establish new product and customer support services. These vendors need to offer to their clients new ICT solutions, easy to be customised and adapted to the SME dynamically changing needs.
The consortium also includes 3 user SMEs (OAS, LESTAR, IMEFY), equipment and automation system providers from different sectors and regions acting at a global market. These SMEs are effectively providing services to their local customers. However, they need ICT based solutions to extend their products by new, cost-effective product and customer support services to be provided to their geographically distributed customers.
Project context and objectives:
Business objectives:
The aim of the project is to provide an innovative comprehensive SW solution, following the Web 2.0 principles, to extend products of automation systems and electrical equipment manufacturer SMEs acting on the global market. Such solution will enable:
(1) ICT and service-engineering vendor SMEs (ICT vendors in the text to follow) to provide a new SW system highly competitive at the ICT market, meeting the urgent needs of their clients / users.
(2) Equipment / component manufacturer SMEs - users of such SW system (users in the text to follow) to strengthen their competitiveness at the global market by providing new product and customer support services and launching new business models.
WEB2SME to support SME needs:
ICT vendor SMEs and service providers are faced with the problem of delivering to their mainly SME clients new SW systems enabling them to cost effectively provide new services to their customers distributed worldwide.
Traditionally, the competitive advantage of many equipment manufacturer SMEs was mostly based in the provision of excellent service to local customers. Presently, industrial SMEs need to operate in a global market. A cost-effective and prompt provision of product and customer support services for geographically distributed customers is a key challenge for globally acting SMEs.
Improved customer support services may allow providing new business models: instead of classical selling of equipment (e.g. control and measurement systems, automation systems, electrical equipment), equipment manufacturer SMEs may overtake guarantee for optimal use of equipment. This in turn will ask for innovative ways to support selection and specification of equipment, monitoring of the equipment usage conditions and functions, including diagnostics and maintenance. This way SMEs will sell not (only) products, but also their knowledge on optimal selection and use of their products, and / or overtake full responsibility for the execution of processes where their equipment is installed. Therefore, such SMEs need to extend their products with different product and customer support services.
Within WEB2SME project extended products deal with areas such as:
(a) condition based maintenance;
(b) problem solving services;
(c) re-configuration of equipment services, etc.
An essential aspect of such PES is that they require a means for efficient collaboration of different actors within a supply chain on one side and customers (end-users of products) on the other side, within a geographically distributed EE context.
Large companies are able to attack the global market by building their own networks worldwide (e.g. subsidiaries, service centres providing local presence in different regions). SMEs, although often being part of different networks, can hardly afford to build highly competent service centres world-wide, having sufficient expertise to efficiently support high variety and permanent evolution of product models and variants. Currently, many SMEs use ICT tools for PES (e.g. for diagnostics, maintenance) which require local presence of SMEs' experts. Therefore, the shift in business models for product extensions presents an important challenge to globally acting SMEs.
The modern ICT-based CWE and knowledge management (KM) technology, offer possibilities to realise product extensions and cost effectively provide product and customer support services to geographically distributed customers by supporting work in virtual world (i.e. over Internet). However, again large companies are the ones that have the means to develop and use complex CWE solutions for collaboration, leaving SMEs behind due to the lack of resources and knowledge they can apply to employ and learn such systems. Building of PES using CWE and KM technology often requires high investments and ICT specialists. Such services have often to be adapted / reconfigured to meet specific and evolving requirements concerning specific products/customers, which imply high, unaffordable costs for SMEs. Therefore, user SMEs refrain from investment in such ICT solutions. This in turn implies a problem for ICT vendor SMEs since most of their potential clients are SMEs. The Web 2.0 principles, aiming to enable users to build / compose SW services by themselves, may offer a promising approach to solve these problems of both equipment users and ICT vendors, i.e. offer possibilities to cost effectively build PES for specific SME needs.
Research and technological development (RTD) needs: WEB2SME intends to provide SW system for globally acting SMEs to effectively build/maintain ICT-based services for product extensions. Although a number of ICT solutions for product extension exist (e.g. for process monitoring, diagnostics, maintenance), they do not meet specific SMEs constraints and do not allow for an efficient 'virtual' collaboration within an SME driven EE, taking into account wide spectrum of dynamically changing needs regarding services, collaboration patterns and different technical backgrounds of the involved actors. The key problem is to enable non-information-technology (IT) experts in SMEs to effectively create / manage services to meet such dynamically changing needs. The effective solution is to apply advanced Web 2.0 principles and service-oriented architecture (SOA) paradigm for CWE in real industrial practice, within an innovative ICT solution affordable for SMEs (low cost, easy-to-integrate). Such CWE solution has to enable cost-effective product and customer support independently of geographical locations of customers and manufacturers. The key RTD challenge, however, is to move from the service composition at individual level of Web 2.0 to company and group level (thus moving from 'mySpace' to 'mySME').
RTD objectives:
The objective is to provide a new ICT solution allowing user SMEs to easily combine application specific SW tools (e.g. 'classical' SW solutions for diagnostics, maintenance) and generic collaborative and KM SW services (addressing application independent functionality to support collaboration, such as discovery of expertise for specific problem, traceability of work, provision of knowledge on products / processes etc.), in order to extend their pro-ducts (equipment, automation systems) by new product and customer support services according to their specific, dynamically changing needs. Such ICT solution should effectively support collaborative work and knowledge sharing among different actors within an EE. It will be open for different information middleware, allowing for remote access over Internet to the equipment / automation system for which PES are provided. To allow for full flexibility and effective building and management of such services by SMEs, the project will provide a set of supporting service engineering tools.
Project results:
Therefore, WEB2SME meets the common business objectives of ICT vendor SMEs, to develop a new ICT system, and industrial user SMEs to extend their products offered at the global market. WEB2SME provides:
(1) An advanced system, based on Web 2.0 principles and open SOA approach, enabling SMEs to cost and time effectively create and manage innovative product and customer support services, allowing flexible collaborative work of different teams (including mobile ones) and customers within international settings, fully adaptive and reconfigurable to specific, dynamically changing, needs of SMEs.
(2) A set of innovative CCS and KM services, which can be combined with application specific (partly existing) tools for product and customer support.
(3) Interfaces to automation systems / equipment (information middleware) and other legacy systems providing (online) data needed for PES.
(4) A set of service engineering tools, enabling SMEs to effectively build new / maintain existing services for product extension i.e. adapt them to the user current needs.
(5) A methodology providing an innovative approach for SMEs in applications of collaborative SW services to realise new product and customer support, based on new forms of collaboration between equipment manufactures and their customers and suppliers.
(6) 3 demonstrators in 3 user SMEs, serving to validate the proposed solutions.
In conclusion the key RTD results is the set of new ICT services and new tools to build innovative PES, especially adapted to the needs of equipment manufacturing SMEs, but applicable in other sectors as well.
Overview of the main components of WEB2SME
1. CCS
CCS represent a generic set of services supporting collaboration amongst teams in an EE. They are intended to enable the user to deal with various tasks (expertise search, collaboration tracking and tracing, etc.) and form the basis upon which to build PES, such as condition-based maintenance or reconfiguration of product. Although these basic services are intentionally generic enough to be employed in a wide range of application scenarios, they offer the possibility for individual customisation to adapt them to specific business needs. The following CCS are implemented within the WEB2SME solution.
(a) resource discovery;
(b) selection of communication services;
(c) collaboration traceability;
(d) team composition;
(e) product / process knowledge provision;
(f) decision support.
However, this list is not exhaustive and additional CCS may be added during later phases of the project if needed and the solution will generally be open for adding new CCS. The following sections describe the basic concept for each of the identified CCS, showing their inputs, outputs, and main functionality.
- Resource discovery: The 'Resource Discovery' service will provide identification of potential collaboration partners for within an established EE context. Potential collaboration partners are discovered based on their possession of a needed expertise as well as location and availability.
- Selection of communication services: This service supports decisions regarding the most appropriate communication service to be used for a team, based on criteria like the personal communication preferences of the team members, the collaboration pattern of the group communication, available communication means.
- Collaboration traceability: The service will be used to record collaborative activities during usage of PES within an extended organisation for later reuse, trying to infer additional knowledge from tracked activities to enhance service functionalities. This way it may be possible to improve recommendations of, e.g. the selection of communication services service based on previous selections made by a user. The service will allow recording of user activities without requiring users to explicitly document their actions while at the same time respecting privacy and intellectual and property right (IPR) issues.
- Team composition: The team composition service will be used for composing a team to achieve a common goal, with members possessing certain needed expertise. In order to achieve this, this service will rely on the Resources Discovery service to find the adequate possible members and compose the team. This team will be specifically adjusted for a particular task (e.g. reconfiguration of a product). Team members will be selected based on their expertise, but their availability and location are decisive factors as well.
- Product / process knowledge provision: This service will provide functionality for searching and retrieving information resources related to products and processes from the WEB2SME solution's common repository as well as other information sources (e.g. automation systems / equipment over information middleware or document management and ERP systems, etc.).
- Decision support: Decision support services will be used to support a user in solving a specific problem in relation to PES, e.g. diagnosing a malfunctioning product. The services may make use of reasoning techniques like rule-based reasoning (RBR) and case-based reasoning (CBR).
2. Services / tools for knowledge sharing
- Document management: Nowadays large quantities of information / knowledge, though available in an electronic form, are stored in an unstructured format (like e.g. documents, pictures, drawings, etc.) as opposed to structured data in a conventional database. This unstructured knowledge is usually intended for processing by human beings and not information systems. The WEB2SME document management service will make this information / knowledge available for processing in different WEB2SME services by applying the three-step process described below:
(a) Monitor: a monitoring component observes systems (e.g. in the most basic form a file system) containing unstructured information for changes.
(b) Parse: content parsers for different data formats offer access to diverse data by trans-forming the original content into a format that can be processed by the following analyser.
(c) Analyse: content analysing services extend the parsed information with further content specific elements, such as title, headings and more, if available. Examples could be table headings of web pages as well as Microsoft Word documents. The parsed and analysed content is further indexed and stored in a repository.
- Monitoring: The monitoring service watches for new or changed files e.g. on a file system on a regular basis. Using file properties (e.g. last modified date) enables the reliable detection of modifications. Changed documents are being queued for analysis and re-monitored to keep the monitoring repository up-to-date. To prevent high system load and network traffic, these monitoring runs may performed in off-peak times. However, this behaviour will be configurable to allow for a more real-time update of the monitoring repository.
- Parsing: The process initiator before analysing and indexing the monitored data is the Parser. The parser is stage two of the three-step process and provides the functionality to access a resource's content and parse the environmental properties of that resource and its system.
An example of specific parser is for a file system. It contains a wide range of different types of data representations. Several come in plain readable ASCII format whereas others are specifically encoded into byte code. Besides the content inside the files other attributes need to be taken into account when parsing the files which can enrich the analysis. Therefore, the content as well as the attributes of files require special handling. Particular parsers for the specific file types offer a larger knowledge output through interpreting instead of plain byte processing and are used to enrich the information. The two parts of a resource that are parsed are the content and the further attributes so-called meta-information of the resource itself. Both need eminent handling, especially the content. Content of resources also occurs in two separate ways: as plain, readable text and byte encoded. Both need different parsing approaches. The byte encoded content needs to be converted in read-able text first, depending on the encoding rules. Afterwards, the ASCII and the translated, byte encoded resource can be further interpreted by configurable rules. Besides the content of resources, specific type properties which are often encoded into the resources (e.g. images contain metadata in special formats) are extracted respectively parsed out of a resource. Like the con-tent, the meta-data encoding differs from type to type and needs the same special handling as the content. Therefore it is part of the type specific parser.
- Analysing: The analysis operation processes the content of the given resources that are handed over by the parsing process. Only resources with accessible content are parsed and further analysed. The content analysis is broken down into the following steps:
(i) extract the content of a resource;
(ii) clean-up the content - removing e.g. 'stop words' that could distort the analysing process;
(iii) analyse the content for domain specific attributes and e.g. textual highlights (e.g. captions);
(iv) send the analysed information to the indexer.
The analysed data is built of the 'stop word' free content. Stop words refer to language specific 'noise' terms that distort the text because of different reasons. Classic stop words in English are 'the', 'of', 'is' and more that have no relevance to the content. Furthermore, these stop words lists can be extended so that domain specific stop words are additionally filtered out of the con-tent. The stop word free content is then indexed and stored in a dedicated repository from which it can be retrieved through specific search functionality.
- Search functionality: The search functionality will provide user interfaces to perform searches for various kinds of knowledge / information to the user. Basically the search functionality will serve as the frontend for the different CCS and services / tools for knowledge sharing that require the ability for a user to search for information, such as the CCS resource discovery or the document management. As such, it is intended to realise the search functionality in a way that allows for customising the base functionality for the specific purposes of the CCS and services / tools for knowledge sharing. This means that the search functionality will provide a basic set of components for defining search criteria as well as presenting and filtering the list of search results. These components are then arranged and combined as needed for the specific search service.
- Reasoning: In order to effectively provide decision support functionality, WEB2SME can rely on existing and proven automated reasoning techniques, like RBR and CBR. Both approaches have their own advantages and disadvantages - as described below - and deployment of either one depends on the target application scenario. How-ever, they appear to offer distinct benefits for realising certain PES, especially for condition-based maintenance, when determining the state of equipment and problem solving when searching for solutions to a problem in equipment.
- RBR: An RBR system consists mainly of three things: facts, rules and an engine that acts on them. Rules represent knowledge and facts represent data. A rule consists of two parts: conditions (if clauses) and actions, where the action part of a rule might assert new facts that fire additional rules. A rule-based system tries to solve problems by applying rules on facts (i.e. matching facts with rules if clauses). RBR systems typically employ one of two reasoning modes: forward chaining or backward chaining. Forward chaining is a data-driven method. Here, the facts in the knowledge base are retained in working memory. The conditional part of the rule is a pattern that must match the facts in order to have the associated action part executed. The inference engine com-pares the facts against patterns and determines which rules are applicable. The engine then selects a rule and executes the actions of the selected rule (the execution may affect the list of applicable rules by adding or removing facts). The inference engine then selects the next applicable rule and executes its actions. This process continues until no applicable rules remain. Backward chaining is a goal driven approach, where the system is presented with a goal and the rules are then checked to see if they can support this goal. If a rule is found whose action part contains the goal the facts in that rule's conditional clause are added as sub-goals and backward-chaining is performed on them recursively. This continues until the sub-goals can be proved by facts in the knowledgebase, in which case the original goal is proved successfully, or until no more rules can fire, in which case the goal remains unproved. In general, backward chaining is more appropriate in deductive systems, where the goal is testing a specific hypothesis. In all other situations, including a case where possible conclusion is not known in advance or a case where several different rules may draw the same conclusion (as it may happen with probabilistic rules), forward chaining is expected to bring better performance.
The last few years brought a renaissance of expert systems, now commonly appearing under the name of rule-based systems or rule-based programming systems. Contrary to classic implementations from the 80s and 90s, which were originally developed as tools for exploration of artificial intelligence, work on new systems has been primarily motivated by desire to separate business logic from its implementation in advanced IT architectures. Thus, the accent in new generation has been on standardisation combined with use of proven algorithms.
- CBR: In contrast to RBR, CBR does not employ pre-defined rules to derive a solution from the given facts but relies on past experiences, i.e. a solution to a given problem is found by matching the current symptoms against the symptoms of problems (cases) that have already been solved in the past. The solution to the best matching case is then considered to be applicable to the problem at hand. The CBR process is formalised into the following four steps:
(a) Retrieve: for the given problem all relevant cases are retrieved from the case base, where a case usually contains a description of the problem and its symptoms as well as the applied solution and possibly some annotations about how the solution was derived.
(b) Reuse: the solution of the best match from the case base is mapped to the current problem. This may also include adapting the found solution to the current needs.
(c) Revise: the (adapted) solution is applied and tested on the real problem. If necessary, i.e. the suggested solution does not prove to successfully solve the problem, the solution needs to be further adapted and refined.
(d) Retain: after having successfully solved the problem, the problem and applied solution are stored in the case base for future reuse.
In comparison to RBR, CBR offers some advantages as well as disadvantages. As advantages one could clearly state that CBR can produce results even if the case base does not contain a 100 % match for the given problem. By employing the concept of similarity, CBR does not require defining explicit rules for any possible condition, whereas a rule-based system will simply not work if none of the rules matches the given facts. Thus, initial set-up is limited to defining the case structure and a way of measuring similarity.
However, depending on the application scenario the task of defining case structure and similarity metric can be equally demanding as covering every possible condition in the rule set. Another advantage is that CBR supports automatic learning as each addition to the case-base will improve results for the next run. A clear disadvantage is the dependence on existing prior knowledge, i.e. without any cases in the case base a CBR system cannot produce any results, similar to a rule-based system when lacking a matching rule for the current set of facts.
Potential impact:
The exploitable results obtained in WEB2SME are:
(1) A system full prototype, as the main project outcome that is targeted at by the involved SMEs. The ICT solution provider related SMEs are specifically aiming to reuse these components after the end of the project to realise further projects with so-called globally acting SMEs to develop solutions that will enable equipment manufacturer SMEs to effectively create / manage PES.
(2) A Methodology full prototype, aiming to provide information about how to introduce in changes / improvements connected with Web 2.0 technology in an efficient manner. It can be used as guidelines for SMEs, which intend to apply the new approach to extend their products based on the WEB2SME services.
The target market for WEB2SME is European ICT industry that has significant manufacturing SME customers, though it can be extended in the future to other sectors and third countries. In terms of business impact the benefits for the SMEs participating in the project WEB2SME are:
- SMEs have now the means to provide high quality and cost effective product and customer support services (improved PES) to geographically distributed customers, leading to increase in competitiveness on a global market and by this to increase in market share (higher turnover and profit). In this sense, each partner SMEs has elaborated their plans on how to use WEB2SME solutions and develop new business models.
- SMEs can reduce their internal costs by introducing higher effectiveness of product and customer support services.
- SMEs can reduce internal costs of SW services since they will be updated by non IT experts, enabled by the Web 2.0 approach.
- SMEs (mainly the ICT vendors) will participate in revenue from selling WEB2SME system.
The dissemination activities carried by the consortium go from internal dissemination activities within the partners to participation in workshops, seminars and conferences or the organisation of user interest groups to make the results of the project available to third parties. The summary of dissemination activities is enumerated below:
- Organisation of 30 internal dissemination activities with different purposes: from requirements gathering to performance of demonstration activities.
- More than 30 individuals and organisations have been informed about the advances of the project after being part of th Users Interest Group.
- Participation in conferences and articles (1. T. Goszczynski, Z. Pilat, M. Slowikowski, J. Zielinski; 'E-business - product and customer support for SME'; In the 6th IPROMS Virtual Conference - 15 - 26 November 2011; T. Goszczynski, Z. Pilat, M. Slowikowski, J. Zielinski; 'Distributed information and communication systems for small and medium Enterprises'; Paper accepted for the AUTOMATION Conference which will be held in Warsaw 6 - 8 April 2011; Article in the business magazine 'Energie-Pflanzen', 17 February 2011; S. Puchalski, Z. Pilat, M. Slowikowski, J. Zielinski; 'Possibility of implementing intelligent welding system supervision using ICT solutions and Web 2.0 services'; Paper for the Automation Conference which will be held in Warsaw 21 - 23 March 2012).
- Participation in 24 workshops in the 3 countries of the project (Poland, Germany and Spain).
- Preparation of a public project presentation, project brochure and web page.
Consortium:
INNOPOLE SL, http://www.innopole.net Spain
ATB Bremen, Institute for Applied Systems Technology Bremen GMBH http://www.atb-bremen.de Germany
- PIAP, Przemyslowy Instytut Automatyki i Pomiarow, http://www.piap.eu Poland
- OAS AG, http://www.oas.de Germany
- ESCO Sp. ZOO, http://www.esco-online.de Poland
- SUPRA ELCO Sp. z.o.o. http://www.supraelco.waw.pl/ Poland
- IMEFY S.L. http://www.imefy.com/ Spain
- freiheit.com GMBH, http://www.freiheit.com Germany.
List of websites: http://www.web2sme.eu
The objective of the project is to provide an innovative comprehensive software (SW) solution, following the Web 2.0 principles, to extend products of automation systems and electrical equipment manufacturer small and medium-sized enterprises (SMEs) acting on the global market.
WEB2SME solution will allow:
(1) industrial user SME - first users of such SW system to strengthen their competitiveness at the global market by providing new product and customer support services and enabling new business models;
(2) information and communication technology (ICT) and service engineering vendor SMEs to provide a new SW system highly competitive at the ICT market.
WEB2SME consortium includes 2 research institutes, 3 ICT vendors and 3 user SMEs, equipment and automation system providers located in three European countries namely Spain, Poland and Germany.
The concept behind WEB2SME is based on the need on user SMEs to extend their products with associated services product extension services (PES) with different product and customer support services, such as condition based maintenance, problem solving, equipment reconfiguration services etc., and to be able to cost effectively provide these services to customers distributed worldwide.
An essential aspect of such PES is that they require a means for efficient collaboration of different actors in a supply chain and customers, within a geographically distributed extended enterprise context. Users need ICT solutions which will allow smooth transition from the current (mostly locally oriented) service provision, to provision of services at global market, partly in 'virtual world'.
The WEB2SME methodology shall serve as guideline for SMEs, which intend to apply the WEB2SME approach - combining collaborative services and KM technology to extend products in SMEs - addressing both technological and organisational aspects related to product extensions and knowledge sharing within the EE in the global market.
Within the project it is planned to realise a comprehensive package of methods, guidelines and manuals which can be divided into the following parts:
(a) methodology to set up and maintain a collaborative working environment (CWE);
(b) methodology to set up new and maintain existing PES; and
(c) manuals.
Within the ICT system concept, the different main components, namely the core collaborative services (CCS) and services / tools for knowledge sharing, will be implemented as well as the concept of PES and how they can be realised based on a combination of core services and different application specific tools.
WEB2SME consortium includes 3 ICT vendor SMEs (INNOPOLE, ESCO, freiheit.com) whose business is to provide novel ICT solutions and support SME users to develop / establish new product and customer support services. These vendors need to offer to their clients new ICT solutions, easy to be customised and adapted to the SME dynamically changing needs.
The consortium also includes 3 user SMEs (OAS, LESTAR, IMEFY), equipment and automation system providers from different sectors and regions acting at a global market. These SMEs are effectively providing services to their local customers. However, they need ICT based solutions to extend their products by new, cost-effective product and customer support services to be provided to their geographically distributed customers.
Project context and objectives:
Business objectives:
The aim of the project is to provide an innovative comprehensive SW solution, following the Web 2.0 principles, to extend products of automation systems and electrical equipment manufacturer SMEs acting on the global market. Such solution will enable:
(1) ICT and service-engineering vendor SMEs (ICT vendors in the text to follow) to provide a new SW system highly competitive at the ICT market, meeting the urgent needs of their clients / users.
(2) Equipment / component manufacturer SMEs - users of such SW system (users in the text to follow) to strengthen their competitiveness at the global market by providing new product and customer support services and launching new business models.
WEB2SME to support SME needs:
ICT vendor SMEs and service providers are faced with the problem of delivering to their mainly SME clients new SW systems enabling them to cost effectively provide new services to their customers distributed worldwide.
Traditionally, the competitive advantage of many equipment manufacturer SMEs was mostly based in the provision of excellent service to local customers. Presently, industrial SMEs need to operate in a global market. A cost-effective and prompt provision of product and customer support services for geographically distributed customers is a key challenge for globally acting SMEs.
Improved customer support services may allow providing new business models: instead of classical selling of equipment (e.g. control and measurement systems, automation systems, electrical equipment), equipment manufacturer SMEs may overtake guarantee for optimal use of equipment. This in turn will ask for innovative ways to support selection and specification of equipment, monitoring of the equipment usage conditions and functions, including diagnostics and maintenance. This way SMEs will sell not (only) products, but also their knowledge on optimal selection and use of their products, and / or overtake full responsibility for the execution of processes where their equipment is installed. Therefore, such SMEs need to extend their products with different product and customer support services.
Within WEB2SME project extended products deal with areas such as:
(a) condition based maintenance;
(b) problem solving services;
(c) re-configuration of equipment services, etc.
An essential aspect of such PES is that they require a means for efficient collaboration of different actors within a supply chain on one side and customers (end-users of products) on the other side, within a geographically distributed EE context.
Large companies are able to attack the global market by building their own networks worldwide (e.g. subsidiaries, service centres providing local presence in different regions). SMEs, although often being part of different networks, can hardly afford to build highly competent service centres world-wide, having sufficient expertise to efficiently support high variety and permanent evolution of product models and variants. Currently, many SMEs use ICT tools for PES (e.g. for diagnostics, maintenance) which require local presence of SMEs' experts. Therefore, the shift in business models for product extensions presents an important challenge to globally acting SMEs.
The modern ICT-based CWE and knowledge management (KM) technology, offer possibilities to realise product extensions and cost effectively provide product and customer support services to geographically distributed customers by supporting work in virtual world (i.e. over Internet). However, again large companies are the ones that have the means to develop and use complex CWE solutions for collaboration, leaving SMEs behind due to the lack of resources and knowledge they can apply to employ and learn such systems. Building of PES using CWE and KM technology often requires high investments and ICT specialists. Such services have often to be adapted / reconfigured to meet specific and evolving requirements concerning specific products/customers, which imply high, unaffordable costs for SMEs. Therefore, user SMEs refrain from investment in such ICT solutions. This in turn implies a problem for ICT vendor SMEs since most of their potential clients are SMEs. The Web 2.0 principles, aiming to enable users to build / compose SW services by themselves, may offer a promising approach to solve these problems of both equipment users and ICT vendors, i.e. offer possibilities to cost effectively build PES for specific SME needs.
Research and technological development (RTD) needs: WEB2SME intends to provide SW system for globally acting SMEs to effectively build/maintain ICT-based services for product extensions. Although a number of ICT solutions for product extension exist (e.g. for process monitoring, diagnostics, maintenance), they do not meet specific SMEs constraints and do not allow for an efficient 'virtual' collaboration within an SME driven EE, taking into account wide spectrum of dynamically changing needs regarding services, collaboration patterns and different technical backgrounds of the involved actors. The key problem is to enable non-information-technology (IT) experts in SMEs to effectively create / manage services to meet such dynamically changing needs. The effective solution is to apply advanced Web 2.0 principles and service-oriented architecture (SOA) paradigm for CWE in real industrial practice, within an innovative ICT solution affordable for SMEs (low cost, easy-to-integrate). Such CWE solution has to enable cost-effective product and customer support independently of geographical locations of customers and manufacturers. The key RTD challenge, however, is to move from the service composition at individual level of Web 2.0 to company and group level (thus moving from 'mySpace' to 'mySME').
RTD objectives:
The objective is to provide a new ICT solution allowing user SMEs to easily combine application specific SW tools (e.g. 'classical' SW solutions for diagnostics, maintenance) and generic collaborative and KM SW services (addressing application independent functionality to support collaboration, such as discovery of expertise for specific problem, traceability of work, provision of knowledge on products / processes etc.), in order to extend their pro-ducts (equipment, automation systems) by new product and customer support services according to their specific, dynamically changing needs. Such ICT solution should effectively support collaborative work and knowledge sharing among different actors within an EE. It will be open for different information middleware, allowing for remote access over Internet to the equipment / automation system for which PES are provided. To allow for full flexibility and effective building and management of such services by SMEs, the project will provide a set of supporting service engineering tools.
Project results:
Therefore, WEB2SME meets the common business objectives of ICT vendor SMEs, to develop a new ICT system, and industrial user SMEs to extend their products offered at the global market. WEB2SME provides:
(1) An advanced system, based on Web 2.0 principles and open SOA approach, enabling SMEs to cost and time effectively create and manage innovative product and customer support services, allowing flexible collaborative work of different teams (including mobile ones) and customers within international settings, fully adaptive and reconfigurable to specific, dynamically changing, needs of SMEs.
(2) A set of innovative CCS and KM services, which can be combined with application specific (partly existing) tools for product and customer support.
(3) Interfaces to automation systems / equipment (information middleware) and other legacy systems providing (online) data needed for PES.
(4) A set of service engineering tools, enabling SMEs to effectively build new / maintain existing services for product extension i.e. adapt them to the user current needs.
(5) A methodology providing an innovative approach for SMEs in applications of collaborative SW services to realise new product and customer support, based on new forms of collaboration between equipment manufactures and their customers and suppliers.
(6) 3 demonstrators in 3 user SMEs, serving to validate the proposed solutions.
In conclusion the key RTD results is the set of new ICT services and new tools to build innovative PES, especially adapted to the needs of equipment manufacturing SMEs, but applicable in other sectors as well.
Overview of the main components of WEB2SME
1. CCS
CCS represent a generic set of services supporting collaboration amongst teams in an EE. They are intended to enable the user to deal with various tasks (expertise search, collaboration tracking and tracing, etc.) and form the basis upon which to build PES, such as condition-based maintenance or reconfiguration of product. Although these basic services are intentionally generic enough to be employed in a wide range of application scenarios, they offer the possibility for individual customisation to adapt them to specific business needs. The following CCS are implemented within the WEB2SME solution.
(a) resource discovery;
(b) selection of communication services;
(c) collaboration traceability;
(d) team composition;
(e) product / process knowledge provision;
(f) decision support.
However, this list is not exhaustive and additional CCS may be added during later phases of the project if needed and the solution will generally be open for adding new CCS. The following sections describe the basic concept for each of the identified CCS, showing their inputs, outputs, and main functionality.
- Resource discovery: The 'Resource Discovery' service will provide identification of potential collaboration partners for within an established EE context. Potential collaboration partners are discovered based on their possession of a needed expertise as well as location and availability.
- Selection of communication services: This service supports decisions regarding the most appropriate communication service to be used for a team, based on criteria like the personal communication preferences of the team members, the collaboration pattern of the group communication, available communication means.
- Collaboration traceability: The service will be used to record collaborative activities during usage of PES within an extended organisation for later reuse, trying to infer additional knowledge from tracked activities to enhance service functionalities. This way it may be possible to improve recommendations of, e.g. the selection of communication services service based on previous selections made by a user. The service will allow recording of user activities without requiring users to explicitly document their actions while at the same time respecting privacy and intellectual and property right (IPR) issues.
- Team composition: The team composition service will be used for composing a team to achieve a common goal, with members possessing certain needed expertise. In order to achieve this, this service will rely on the Resources Discovery service to find the adequate possible members and compose the team. This team will be specifically adjusted for a particular task (e.g. reconfiguration of a product). Team members will be selected based on their expertise, but their availability and location are decisive factors as well.
- Product / process knowledge provision: This service will provide functionality for searching and retrieving information resources related to products and processes from the WEB2SME solution's common repository as well as other information sources (e.g. automation systems / equipment over information middleware or document management and ERP systems, etc.).
- Decision support: Decision support services will be used to support a user in solving a specific problem in relation to PES, e.g. diagnosing a malfunctioning product. The services may make use of reasoning techniques like rule-based reasoning (RBR) and case-based reasoning (CBR).
2. Services / tools for knowledge sharing
- Document management: Nowadays large quantities of information / knowledge, though available in an electronic form, are stored in an unstructured format (like e.g. documents, pictures, drawings, etc.) as opposed to structured data in a conventional database. This unstructured knowledge is usually intended for processing by human beings and not information systems. The WEB2SME document management service will make this information / knowledge available for processing in different WEB2SME services by applying the three-step process described below:
(a) Monitor: a monitoring component observes systems (e.g. in the most basic form a file system) containing unstructured information for changes.
(b) Parse: content parsers for different data formats offer access to diverse data by trans-forming the original content into a format that can be processed by the following analyser.
(c) Analyse: content analysing services extend the parsed information with further content specific elements, such as title, headings and more, if available. Examples could be table headings of web pages as well as Microsoft Word documents. The parsed and analysed content is further indexed and stored in a repository.
- Monitoring: The monitoring service watches for new or changed files e.g. on a file system on a regular basis. Using file properties (e.g. last modified date) enables the reliable detection of modifications. Changed documents are being queued for analysis and re-monitored to keep the monitoring repository up-to-date. To prevent high system load and network traffic, these monitoring runs may performed in off-peak times. However, this behaviour will be configurable to allow for a more real-time update of the monitoring repository.
- Parsing: The process initiator before analysing and indexing the monitored data is the Parser. The parser is stage two of the three-step process and provides the functionality to access a resource's content and parse the environmental properties of that resource and its system.
An example of specific parser is for a file system. It contains a wide range of different types of data representations. Several come in plain readable ASCII format whereas others are specifically encoded into byte code. Besides the content inside the files other attributes need to be taken into account when parsing the files which can enrich the analysis. Therefore, the content as well as the attributes of files require special handling. Particular parsers for the specific file types offer a larger knowledge output through interpreting instead of plain byte processing and are used to enrich the information. The two parts of a resource that are parsed are the content and the further attributes so-called meta-information of the resource itself. Both need eminent handling, especially the content. Content of resources also occurs in two separate ways: as plain, readable text and byte encoded. Both need different parsing approaches. The byte encoded content needs to be converted in read-able text first, depending on the encoding rules. Afterwards, the ASCII and the translated, byte encoded resource can be further interpreted by configurable rules. Besides the content of resources, specific type properties which are often encoded into the resources (e.g. images contain metadata in special formats) are extracted respectively parsed out of a resource. Like the con-tent, the meta-data encoding differs from type to type and needs the same special handling as the content. Therefore it is part of the type specific parser.
- Analysing: The analysis operation processes the content of the given resources that are handed over by the parsing process. Only resources with accessible content are parsed and further analysed. The content analysis is broken down into the following steps:
(i) extract the content of a resource;
(ii) clean-up the content - removing e.g. 'stop words' that could distort the analysing process;
(iii) analyse the content for domain specific attributes and e.g. textual highlights (e.g. captions);
(iv) send the analysed information to the indexer.
The analysed data is built of the 'stop word' free content. Stop words refer to language specific 'noise' terms that distort the text because of different reasons. Classic stop words in English are 'the', 'of', 'is' and more that have no relevance to the content. Furthermore, these stop words lists can be extended so that domain specific stop words are additionally filtered out of the con-tent. The stop word free content is then indexed and stored in a dedicated repository from which it can be retrieved through specific search functionality.
- Search functionality: The search functionality will provide user interfaces to perform searches for various kinds of knowledge / information to the user. Basically the search functionality will serve as the frontend for the different CCS and services / tools for knowledge sharing that require the ability for a user to search for information, such as the CCS resource discovery or the document management. As such, it is intended to realise the search functionality in a way that allows for customising the base functionality for the specific purposes of the CCS and services / tools for knowledge sharing. This means that the search functionality will provide a basic set of components for defining search criteria as well as presenting and filtering the list of search results. These components are then arranged and combined as needed for the specific search service.
- Reasoning: In order to effectively provide decision support functionality, WEB2SME can rely on existing and proven automated reasoning techniques, like RBR and CBR. Both approaches have their own advantages and disadvantages - as described below - and deployment of either one depends on the target application scenario. How-ever, they appear to offer distinct benefits for realising certain PES, especially for condition-based maintenance, when determining the state of equipment and problem solving when searching for solutions to a problem in equipment.
- RBR: An RBR system consists mainly of three things: facts, rules and an engine that acts on them. Rules represent knowledge and facts represent data. A rule consists of two parts: conditions (if clauses) and actions, where the action part of a rule might assert new facts that fire additional rules. A rule-based system tries to solve problems by applying rules on facts (i.e. matching facts with rules if clauses). RBR systems typically employ one of two reasoning modes: forward chaining or backward chaining. Forward chaining is a data-driven method. Here, the facts in the knowledge base are retained in working memory. The conditional part of the rule is a pattern that must match the facts in order to have the associated action part executed. The inference engine com-pares the facts against patterns and determines which rules are applicable. The engine then selects a rule and executes the actions of the selected rule (the execution may affect the list of applicable rules by adding or removing facts). The inference engine then selects the next applicable rule and executes its actions. This process continues until no applicable rules remain. Backward chaining is a goal driven approach, where the system is presented with a goal and the rules are then checked to see if they can support this goal. If a rule is found whose action part contains the goal the facts in that rule's conditional clause are added as sub-goals and backward-chaining is performed on them recursively. This continues until the sub-goals can be proved by facts in the knowledgebase, in which case the original goal is proved successfully, or until no more rules can fire, in which case the goal remains unproved. In general, backward chaining is more appropriate in deductive systems, where the goal is testing a specific hypothesis. In all other situations, including a case where possible conclusion is not known in advance or a case where several different rules may draw the same conclusion (as it may happen with probabilistic rules), forward chaining is expected to bring better performance.
The last few years brought a renaissance of expert systems, now commonly appearing under the name of rule-based systems or rule-based programming systems. Contrary to classic implementations from the 80s and 90s, which were originally developed as tools for exploration of artificial intelligence, work on new systems has been primarily motivated by desire to separate business logic from its implementation in advanced IT architectures. Thus, the accent in new generation has been on standardisation combined with use of proven algorithms.
- CBR: In contrast to RBR, CBR does not employ pre-defined rules to derive a solution from the given facts but relies on past experiences, i.e. a solution to a given problem is found by matching the current symptoms against the symptoms of problems (cases) that have already been solved in the past. The solution to the best matching case is then considered to be applicable to the problem at hand. The CBR process is formalised into the following four steps:
(a) Retrieve: for the given problem all relevant cases are retrieved from the case base, where a case usually contains a description of the problem and its symptoms as well as the applied solution and possibly some annotations about how the solution was derived.
(b) Reuse: the solution of the best match from the case base is mapped to the current problem. This may also include adapting the found solution to the current needs.
(c) Revise: the (adapted) solution is applied and tested on the real problem. If necessary, i.e. the suggested solution does not prove to successfully solve the problem, the solution needs to be further adapted and refined.
(d) Retain: after having successfully solved the problem, the problem and applied solution are stored in the case base for future reuse.
In comparison to RBR, CBR offers some advantages as well as disadvantages. As advantages one could clearly state that CBR can produce results even if the case base does not contain a 100 % match for the given problem. By employing the concept of similarity, CBR does not require defining explicit rules for any possible condition, whereas a rule-based system will simply not work if none of the rules matches the given facts. Thus, initial set-up is limited to defining the case structure and a way of measuring similarity.
However, depending on the application scenario the task of defining case structure and similarity metric can be equally demanding as covering every possible condition in the rule set. Another advantage is that CBR supports automatic learning as each addition to the case-base will improve results for the next run. A clear disadvantage is the dependence on existing prior knowledge, i.e. without any cases in the case base a CBR system cannot produce any results, similar to a rule-based system when lacking a matching rule for the current set of facts.
Potential impact:
The exploitable results obtained in WEB2SME are:
(1) A system full prototype, as the main project outcome that is targeted at by the involved SMEs. The ICT solution provider related SMEs are specifically aiming to reuse these components after the end of the project to realise further projects with so-called globally acting SMEs to develop solutions that will enable equipment manufacturer SMEs to effectively create / manage PES.
(2) A Methodology full prototype, aiming to provide information about how to introduce in changes / improvements connected with Web 2.0 technology in an efficient manner. It can be used as guidelines for SMEs, which intend to apply the new approach to extend their products based on the WEB2SME services.
The target market for WEB2SME is European ICT industry that has significant manufacturing SME customers, though it can be extended in the future to other sectors and third countries. In terms of business impact the benefits for the SMEs participating in the project WEB2SME are:
- SMEs have now the means to provide high quality and cost effective product and customer support services (improved PES) to geographically distributed customers, leading to increase in competitiveness on a global market and by this to increase in market share (higher turnover and profit). In this sense, each partner SMEs has elaborated their plans on how to use WEB2SME solutions and develop new business models.
- SMEs can reduce their internal costs by introducing higher effectiveness of product and customer support services.
- SMEs can reduce internal costs of SW services since they will be updated by non IT experts, enabled by the Web 2.0 approach.
- SMEs (mainly the ICT vendors) will participate in revenue from selling WEB2SME system.
The dissemination activities carried by the consortium go from internal dissemination activities within the partners to participation in workshops, seminars and conferences or the organisation of user interest groups to make the results of the project available to third parties. The summary of dissemination activities is enumerated below:
- Organisation of 30 internal dissemination activities with different purposes: from requirements gathering to performance of demonstration activities.
- More than 30 individuals and organisations have been informed about the advances of the project after being part of th Users Interest Group.
- Participation in conferences and articles (1. T. Goszczynski, Z. Pilat, M. Slowikowski, J. Zielinski; 'E-business - product and customer support for SME'; In the 6th IPROMS Virtual Conference - 15 - 26 November 2011; T. Goszczynski, Z. Pilat, M. Slowikowski, J. Zielinski; 'Distributed information and communication systems for small and medium Enterprises'; Paper accepted for the AUTOMATION Conference which will be held in Warsaw 6 - 8 April 2011; Article in the business magazine 'Energie-Pflanzen', 17 February 2011; S. Puchalski, Z. Pilat, M. Slowikowski, J. Zielinski; 'Possibility of implementing intelligent welding system supervision using ICT solutions and Web 2.0 services'; Paper for the Automation Conference which will be held in Warsaw 21 - 23 March 2012).
- Participation in 24 workshops in the 3 countries of the project (Poland, Germany and Spain).
- Preparation of a public project presentation, project brochure and web page.
Consortium:
INNOPOLE SL, http://www.innopole.net Spain
ATB Bremen, Institute for Applied Systems Technology Bremen GMBH http://www.atb-bremen.de Germany
- PIAP, Przemyslowy Instytut Automatyki i Pomiarow, http://www.piap.eu Poland
- OAS AG, http://www.oas.de Germany
- ESCO Sp. ZOO, http://www.esco-online.de Poland
- SUPRA ELCO Sp. z.o.o. http://www.supraelco.waw.pl/ Poland
- IMEFY S.L. http://www.imefy.com/ Spain
- freiheit.com GMBH, http://www.freiheit.com Germany.
List of websites: http://www.web2sme.eu