Thesis Diseño de un modelo de negocios para una startup tecnológica que desarrolla un software basado en inteligencia artificial
Loading...
Date
2020-12-28
Journal Title
Journal ISSN
Volume Title
Program
Ingeniería Comercial
Departament
Campus
Campus Casa Central Valparaíso
Abstract
El presente trabajo de titulación tiene como objetivo el diseño de un modelo de negocios para una startup que desarrolla un software basado en Inteligencia Artificial con técnicas de Procesamiento del Lenguaje Natural, que busca mejorar y aumentar la eficiencia operativa en el proceso de otorgamiento de créditos y convenios, y su respectivo análisis de políticas y normativa crediticia en instituciones financieras, como cooperativas de ahorro y crédito y la banca.
Para ello, la metodología utilizada involucró un proceso de revisión de distintas fuentes bibliografía para conocer aspectos relevantes del sector bancario y las cooperativas de ahorro y crédito, las nuevas tecnologías relacionadas con la Inteligencia Artificial y las actuales soluciones propuestas al sector financiero. A su vez, se realizó una investigación de campo a través de entrevistas en profundidad a profesionales del área bancaria y a académico experto.
Por otra parte, se llevó a cabo la construcción del modelo de negocios, aplicando la metodología Canvas de Alexander Osterwalder e Yves Pigneaur, donde se identificaron los procesos y actividades primordiales para entregar a los potenciales clientes, una propuesta de valor sustentada en la eficiencia operativa, automatización de tareas y trabajo como experto.
Finalmente, se realizó un estudio económico-financiero, que considera un horizonte de evaluación de 5 años de operación del proyecto, con una tasa de descuento del 4,88% a partir del método CAPM, arrojando un VAN positivo de 27956,66 UF, una Tasa Interna de Retorno de 73% y recuperación de la inversión al quinto año de operaciones.
Los resultados de sensibilidad y riesgo muestran que las variables críticas con mayor impacto corresponden a la cantidad de licencias adquiridas por los clientes, la tasa de crecimiento anual de clientes y el precio de las suscripciones. Por lo tanto, en la medida que se mitigue el riesgo de tales variables integrando otras herramientas complementarias a la plataforma, se podrá entregar una propuesta atractiva para los potenciales clientes del sector financiero.
The objective of this paper is to design a business model for a startup that develops a software based on Artificial Intelligence with Natural Language Processing techniques, which seeks to improve and increase operational efficiency in the process of granting loans and agreements, and their respective analysis of credit policies and regulations in financial institutions, such as credit unions and banks. To this end, the methodology used involved a process of reviewing different bibliographic sources to learn about relevant aspects of the banking sector and credit unions, new technologies related to Artificial Intelligence and current solutions proposed to the financial sector. At the same time, field research was carried out through in-depth interviews with professionals from the banking area and academic experts. On the other hand, the construction of the business model was carried out, applying the Canvas methodology of Alexander Osterwalder and Yves Pigneaur, where the processes and primordial activities were identified to deliver to the potential clients, a proposal of value sustained in the operative efficiency, automation of tasks and work as an expert. Finally, an economic-financial study was carried out, which considers an evaluation horizon of 5 years of project operation, with a discount rate of 4.88% based on the CAPM method, yielding a positive NPV of 27956.66 UF, an Internal Rate of Return of 73% and a recovery of the investment in the fifth year of operations. The sensitivity and risk results show that the critical variables with the greatest impact correspond to the number of licenses acquired by clients, the annual growth rate of clients and the price of subscriptions. Therefore, to the extent that the risk of such variables is mitigated by integrating other complementary tools to the platform, an attractive proposal for potential clients in the financial sector can be delivered.
The objective of this paper is to design a business model for a startup that develops a software based on Artificial Intelligence with Natural Language Processing techniques, which seeks to improve and increase operational efficiency in the process of granting loans and agreements, and their respective analysis of credit policies and regulations in financial institutions, such as credit unions and banks. To this end, the methodology used involved a process of reviewing different bibliographic sources to learn about relevant aspects of the banking sector and credit unions, new technologies related to Artificial Intelligence and current solutions proposed to the financial sector. At the same time, field research was carried out through in-depth interviews with professionals from the banking area and academic experts. On the other hand, the construction of the business model was carried out, applying the Canvas methodology of Alexander Osterwalder and Yves Pigneaur, where the processes and primordial activities were identified to deliver to the potential clients, a proposal of value sustained in the operative efficiency, automation of tasks and work as an expert. Finally, an economic-financial study was carried out, which considers an evaluation horizon of 5 years of project operation, with a discount rate of 4.88% based on the CAPM method, yielding a positive NPV of 27956.66 UF, an Internal Rate of Return of 73% and a recovery of the investment in the fifth year of operations. The sensitivity and risk results show that the critical variables with the greatest impact correspond to the number of licenses acquired by clients, the annual growth rate of clients and the price of subscriptions. Therefore, to the extent that the risk of such variables is mitigated by integrating other complementary tools to the platform, an attractive proposal for potential clients in the financial sector can be delivered.
Description
Keywords
Instituciones financieras, Inteligencia artificial, Procesamiento del lenguaje
