الفهرس | Only 14 pages are availabe for public view |
Abstract Nowadays, manufacturers are shifting from a traditional product-centric business paradigm to a service-centric one by offering not only products but products accompanied by services, which is known as Product-Service Systems (PSSs). PSS mass customization entails configuring products with varying degrees of differentiation to meet the needs of various customers. This is combined with service customization, in which configured products are expanded by customers to include smart IoT devices (e.g., sensors) to improve product usage and facilitate the transition to smart connected products. A massive amount of data is collected along the PSS customization lifecycle starting from the early stages of smart product ideation and customization to smart product monitoring and improvement. This data is useless and invaluable unless it is used to generate more information and gain insights. Moreover, this massive data overload issue hinders the various stakeholders involved in the PSS customization lifecycle from making informed decisions. However, PSSs do not support the analysis of this collected data to enhance data-driven decision-making. This creates a demand for the adoption of novel techniques/approaches to assist all the involved stakeholders in making informed decisions and accelerating the different PSS customization lifecycle processes. We anticipate that data analytics techniques can be utilized to analyze the massive amounts of data collected during the PSS customization lifecycle. Data analytics techniques are classified into three categories: descriptive, predictive, and prescriptive. Recommender Systems (RSs) fall under the bigger class of prescriptive analytics, which represent software tools that offer better suggestions to customers, taking into account their requirements/preferences. We anticipate that RSs could play a pivotal role in the different processes of the PSS customization lifecycle (i.e., smart product ideation, PSS customization, production planning, production execution, and production monitoring) by assisting various involved stakeholders in making informed right decisions. Accordingly, in this thesis, a recommendation framework is proposed to support the different processes of the PSS customization lifecycle. In this framework, a set of recommendation capabilities are identified to support and accelerate the different processes of the PSS customization lifecycle while accommodating different stakeholders’ perspectives. Then, we concentrated our efforts on addressing the main problems identified for the smart product ideation and the customization of services as part of the PSS customization lifecycle, which are as follows, (i) in the smart product ideation process, customers may start their customization process by selecting a PSS variant from a wide range of available previously customized PSS variants rather than doing customization from scratch. Consequently, finding a PSS variant that is precisely aligned to the customers’ requirements is a cognitive task that the customers are unable to manage easily; and (ii) during the service customization process, customers are interested in expanding existing configured products to include smart sensors or IoT communication devices in general, to improve product usage and facilitate the transition to smart connected products. Despite the significant gained value from adding sensors to products, the selection of the appropriate types of sensors and their adequate locations is a challenge that customers are unable to manage easily and effectively. These problems are addressed by proposing recommendation approaches that assist customers in making informed decisions during the previously mentioned two processes. For the smart product ideation process, we propose a hybrid knowledge-based recommendation approach that assists customers in selecting a previously customized PSS variant that is accurately aligned to their requirements from a wide range of available ones. The proposed approach models the problem of selecting previously customized PSS variants as a Constraint Satisfaction Problem (CSP), to filter out PSS variants that do not satisfy customers’ needs. After that, a weighted utility function is applied to rank the remaining PSS variants based on their utility to the customer. The utility and applicability of the proposed recommendation approach for the smart product ideation process is demonstrated through its application on a real-life case study in the domain of laser machines. Moreover, the proposed approach is evaluated through feedback from industrial experts. The evaluation results show the effectiveness, utility, efficiency, and persuasiveness of our proposed approach. In addition, for the customization of services process, we propose a data warehouse-based recommendation approach that assists customers in selecting the appropriate types of smart devices (e.g., sensors) to install on their configured products and their adequate locations. This approach collects and analyzes usage incident data generated during the usage phase of similar products to the one that the target customer wishes to expand by adding smart sensors. The analysis of this data helps in identifying the most critical parts with the highest number of incidents, the causes of these incidents, and the neighboring influential parts that are responsible for the occurrence of these incidents that occurred on those critical parts. As a result, these critical parts are suggested to the target customer as the most important parts to where sensors should be installed in her current product. A real-life case study in the domain of milling machines is used to demonstrate the utility and applicability of the proposed approach for the customization of services process. Moreover, the performance of the proposed approach is evaluated in terms of response time. The evaluation results show that our proposed approach is able to generate recommendations within the recommended system response time boundaries. |