AI4INVENTORY

PROJECT TITLE:

AI4INVENTORY

GENERAL DATA:

Program: support grants for Innovative Business Groupings to improve the competitiveness of small and medium-sized companies and the call for applications for the year 2022, within the framework of the Recovery, Transformation and Resilience Plan.
Referencia: AEI-010500-2022b-234
RESOLUTION Date: 11/23/2022
Execution date: 06/30/2022 – 04/29/2023

PRESENTATION AND OBJECTIVES:

Today, we are living in complex times for industry and in particular for machine and component manufacturers, who are increasingly dependent on complex global supply chains. Uncertainty seems to be here to stay for a good season and the current crisis, initially triggered by the global pandemic and accentuated by the war in Ukraine, has caused a situation of stress to the limit in many sectors.
Every morning, industrial companies are faced with the uncertainty of the impact that this context of crisis will have on transportation, procurement, material costs and production processes in general. Companies are continuously replanning, adapting their production rhythms to the daily situation and sizing, as far as possible, their production capacities and stocks to the rhythms of demand and the availability of their raw materials.
With this, and in a global approach such as the one we have, competition between companies is not so much about their products, but about their capacity for resilience, for responding to the volatility of demand and for adapting their supply chains. The company with the most efficient supply chain will be in a better position to compete because it will be able to minimize total costs without reducing the service levels demanded by customers.
However, in the current situation, the need for proper inventory and procurement planning has become even more pronounced. Given transportation problems, credit shortages, rising costs, or the need not to stop production processes, companies need to reduce their investment in working capital and optimize their stock in a more profitable way, with a more appropriate mix of products and raw materials in the warehouse, according to the stock-service strategy selected by the company. It must be taken into account that the purchasing process in machinery manufacturing companies is not simple since many criteria converge when choosing one or another supplier. In addition, purchases have a major impact on your bottom line. In fact, in many companies in the sector, purchases exceed 50% of the expenses in the profit and loss account.
How can these processes be improved? It can be improved through proper modeling of demand and supply uncertainty for each product, and through the application of demand forecasting and optimization technologies on the product set. Inventory optimization based on the automatic processing of thousands of SKUs allows balancing capital investment and service level objectives, thus improving the company’s profitability.
And this is what this project pursues, betting on the most innovative techniques in the area of Artificial Intelligence and Machine Learning through an automated analysis platform, which allows optimizing the processes of demand forecasting, inventory management and supply in an integrated way, with the aim of cushioning uncertainty and being prepared to anticipate and adapt supply models in an agile way. Developing a good procurement planning system adapted to the specifics of the company ensures stock availability for customers while minimizing inventory holding costs.

Other beneficial derivatives of this type of systems are the following:

• Increase visibility into the supply chain by estimating demand in order to anticipate complexities and make better decisions.
• Sizing the inventory of each reference in the warehouse to guarantee the required service level at the minimum cost.
• Provide daily order planning with a time horizon of one month and with a daily update frequency.
• Reduce planning cycle time by automating the process, in a management by exception framework.
• Systematize the supply planning process through a single process of demand forecasting, inventory sizing and order planning.
• Determine the target service level for each component in stock in such a way that the service level of the final product can be maximized with a given limit of investment in stock.

For these reasons, and in order to accelerate their immersion and adaptation to the precepts of Industry 4.0, it is strategic that construction machinery manufacturers begin to incorporate Artificial Intelligence and Machine Learning technologies to help them make purchasing decisions regarding the selection of the best suppliers.

The project aims to introduce innovative techniques in the area of Artificial Intelligence and Machine Learning for the optimization of demand forecasting, inventory management and procurement processes in an integrated manner, with the objective of buffering uncertainty, anticipating problems in this field and adapting supply models in an agile manner.

The project involves 5 SME companies with different problems in this field and will allow validation of the usefulness of these technologies in different application cases and with different objectives within the general theme of demand forecasting, inventory management and procurement.
The participating SMEs are mainly machinery manufacturers and component suppliers, considering that these companies are facing important problems in these aspects, especially in the current times of great uncertainty in the supply chain and the need to combine production-to-order models with inventory procurement models for production and spare parts.

This problem occurs in different sectors with specific particularities in each one of them. Hence the convenience of integrating companies from two different AEIs, CAMPAG and AERA, into the project.

This approach is aimed at the participating AEIs and their sectors:

• Introduce advanced technologies in the field of demand management, inventories and procurement, processes that have traditionally been carried out in a “manual” and non-systematized manner, thus increasing the degree of digitalization and modernization of companies.
• To make an important efficiency leap in these processes that could serve as a model for other companies in the participating AEIs to consider moving forward along the same path.
• Encourage collaboration with technology centers and break down barriers and reluctance to access technologies such as artificial intelligence and machine learning, which are often considered beyond the reach and understanding of SMEs.
• Encourage collaboration among the participating companies so that they can take reference points for best practices in these processes.
  Technological objectives

At the technological level, the following objectives are proposed:

• Evaluate different methods for demand analysis and forecasting in order to identify those best suited to the specific problems of the sectors addressed and to the particularities of their production and management models.
• Evaluate new multivariate methods and specialized methods for low turnover and intermittent demand.
• Evaluate various uncertainty functions and long tail inventory models that can occur in these cases.

PARTICIPATING ENTITIES:

INDUSTRIAS MANRIQUE SA
MAQUINARIA AGRICOLA Y ACCESORIOS SL
HINE ZARAGOZA SL
SIJALON SL
BMC MAQUINARIA AGRICOLA SL
ARAGONESE AERONAUTICAL ASSOCIATION
ARAGON AGRICULTURAL MACHINERY CLUSTER
TECHNOLOGICAL INSTITUTE OF ARAGON

BUDGET:

TOTAL PROJECT BUDGET: 199,692 €.
TOTAL BUDGET ITAINNOVA: 147,432 €.

FINANCING:

TOTAL PROJECT FINANCING: 82,110 €.
TOTAL FINANCING ITANNOVA: 53,371 €.
This project is funded by the Ministry of Industry, Trade and Tourism of the Government of Spain, under the support program for innovative business clusters and by the European Union “Next GeneratinEU” PRTR, being project reference AEI-010500-2022b-234.