The AITrawel Project
Let’s explore how AI integration enhances simulation models, unlocking untapped potential in supply chain optimization. A game-changer for decision-making in logistics and beyond with STAM.
Categories
- Please introduce your company in one sentence.
STAM is an SME based in Genoa, Italy, that offers high-tech services in various areas of engineering. The company is very active in research and is among the top Italian SMEs for projects funded in the European framework. - Kindly provide a one-paragraph description of your project.
AITrawel represents an opportunity for STAM to strengthen and enrich its know-how and offering in the area of logistics and supply-chain. One of the most sensitive topics for STAM’s customers is certainly the optimisation of transports, taking into account various performance indicators and the demand variability. Over the years, STAM specialized in developing a solid knowledge in developing digital twins and simulators for analysing and testing real supply-chain performances and dynamics in virtual environments. These simulation models exploit and combine various modelling approaches (including system dynamics, agent-based and discrete event) to faithfully reproduce the specific case study.By means of these tools, it is possible to reproduce the functioning of the different processes and phases that characterise the supply chain, taking into account the existing relationships and also the constraints determined, for example, by capacity and space. Themodels are tailored to the customer’s needs so that they can function as decision support tools. In fact, the user can generate so-called what-if scenarios with which the performance of the system under different situations and boundary conditions can be simulated. Although this technology is very high-performant and functional for the achievement of objectives, the inherent potential of the volume of data generated by simulation campaigns is not actually exploited to the full. So, STAM considers AITrawel as a key opportunity to implement strong AI-based components in its simulation models using the simulation results to feed and train the AI-based algorithms. - How is the StairwAI support program contributing to the development of your solution?
STAM has a low level of AI skills in the company. Under StairwAI, STAM intends to increase its skill set and implement an AI-based module that can receive and process the huge dataset of simulation outputs in a short time to obtain more reliable and understandable results for customers. AITrawel would allow STAM to halve the effort traditionally required to analyze simulation results, while offering its customers a greatly improved tool from a performance perspective. Indeed, the system modeling phase itself is very resource- and time-consuming; however, the phase of analyzing and interpreting the results also requires similar effort.Therefore, the ability to facilitate the data analysis process to return synthetic results would be of great use for the commercial purposes of STAM. - Could you outline the upcoming roadmap or plan for the next few months?
AITrawel will be developed over the course of six months of activities. In addition to the first two months, which are more focused on the feasibility study phase, technical activities have been identified and organized into 4 different phases:1. Requirements gathering; for which a reference case study will be selected, including for final testing of the solution. In relation to the collected requirements, the objectives will be defined and consequently the technical aspects to be investigated in the following phases will be outlined.2. Model development; in this phase, case study modeling activities will be carried out, creating a fully parametric simulator, in which input variables can be can be set to generate different “what-if” scenarios.3. Integration AI-based module; in parallel, the technical support best suited to the ambitions and requirements of the project will be identified and integrated into the simulator.4. Testing and validation; the last phase is the collection and processing of results to test and validate the system.
Let’s see how the implementation of AI can improve their solution!