How Machine Learning Can Be Applied in Demand Forecasting and Supply Planning
Demand forecasting is known to be challenging due to phenomena such as the Bullwhip Effect (Lee et al., 1997a). Having worked in supply chain management for many years, I observed that the Bullwhip Effect is a universal issue across all industries. Companies, regardless of size, struggle with the amplification and distortion of demand information. My master’s thesis focused on leveraging AI, specifically machine learning, to enhance demand forecasting and optimize supply chains for the case company, a producer of durable IT devices. This thesis aimed to bridge two hot topics in the digital era — machine learning and demand forecasting — by providing a practical solution within a real business context. Effective demand forecasting and supply planning are crucial components of supply chain management. Inaccurate demand information often leads to suboptimal decisions, causing inventory imbalances and customer dissatisfaction. Many organizations struggle with business challenges due to inaccurate demand forecasts, resulting in inefficiencies, financial losses, and unhappy customers. Recent advancements in machine learning (ML) algorithms offer new tools to improve forecasting accuracy and maintain excellent performance for industrial demand. Since 2018, Machine Learning algorithms have consistently won competitions focused on retail demand forecasting. In my study, I employed an applied action research approach to diagnose the case company's challenges and offer viable solutions. Data collection primarily involved qualitative methods such as interviews, meetings, and internal document analysis, supplemented by quantitative data for model development. Four algorithms were used to build Machine Learning models using data extracted from the company's weekly demand reports: Linear Regression, Decision Tree, Recurrent Neural Network, and Support Vector Machine (as recommended by: Vandeput, 2023). After processing the data, performing feature engineering, and conducting training, testing, and validation, Linear Regression emerged as the most suitable algorithm based on both Machine Learning metrics and internal evaluations. Based on the thesis results, this approach was recommended for integration into the case company's existing demand forecasting and supply planning processes to support better decision-making. The steps in the approach are shown in Figure 1 below. Figure 1. Demand Forecasting Process by Utilizing Machine Learning Model (Zhang 2024). By leveraging AI technology, particularly machine learning, in demand forecasting and supply planning, organizations can vastly improve supply chain management. Enhanced forecast accuracy and optimized inventory levels lead to reduced inefficiencies, minimized financial losses, and heightened customer satisfaction. My academic journey in Metropolia University of Applied Sciences has equipped me with invaluable expertise in digitalization and data analytics. Notably, Power BI and Machine Learning provided direct insights for visualizing data, aiding in the selection of the most suitable machine learning model in my thesis. About the author Yi Zhang is a graduate of Master´s degree program in Business Informatics. Yi is keen on new technologies and has set her personal goal to master Machine Learning for advancing in her professional area. References Lee, H.L., Padmanabhan, V. & Whang, S. (1997 a). The Bullwhip Effect in Supply Chains. Sloan Management Review/Spring. Vandeput, N. (2023). Demand forecasting best practices. Manning Publications. Available from: https://learning.oreilly.com/library/view/demand-forecasting-best/9781633438095/?sso_link=yes&sso_link_from=metropolia-university Zhang, Y. (2024). Machine Learning Applied in Demand Forecasting and Supply Planning. Metropolia UAS, Master´s thesis, 99 pages. Available from: https://www.theseus.fi/handle/10024/857760
Integrating Security into Continuous Delivery
Adopting a DevOps strategy for software development aims to significantly increase the speed of software delivery process by working in small batches and ensuring software is always releasable. This way of working is often called Continuous Delivery. However, the increased speed in software delivery creates challenges for existing security processes and practices. To ensure security concerns are identified before the software is released, security must be integrated into the Continuous Delivery process. This was the topic of my Master’s thesis that has just been completed as part of Metropolia Master´s studies. When working as a consultant helping organizations with all things around DevOps and Continuous Delivery, I have noticed that security is still often not integrated into the process as well as it could. For sure, most professionals try to think of security while implementing new features and automation. Yet, often I think we tell ourselves that there should be security experts in the organization that will be ultimately responsible for the security of the solution. Here, I can take myself as an example. Although I have discovered many great open-source security tools that could be used at various stages of the software development and delivery process, rarely did I feel like I have the time and mandate to take them into real use in projects. This is a mindset which, I think, should be changed; everyone who contributes to the software delivery should be responsible for security. Realizing it as a problem, I turned this challenge into a Master’s Thesis topic when I started my studies at Metropolia. According to my initial idea, integrating the available open-source security tools into the Continuous Delivery processes would provide a fast feedback loop on security threats and vulnerabilities that developers might accidently introduce while working on projects (Vainio 2023). This is what the Master´s thesis finally achieved. What is Continuous Delivery? In my experience, a successful DevOps strategy for software delivery revolves around the concept of Continuous Delivery which was popularized by David Farley and Jez Humble in their 2010 book called “Continuous Delivery”. (Farley & Humble 2010) More technical readers will know that continuous Delivery extends the earlier coined Continuous Integration concept and takes it to its logical conclusion: every change to the software should be followed by multiple stages of automated testing to verify that the software is releasable; and if the testing fails, everyone must work together to either rollback the change or fix the issue. These stages are arranged into what is called the deployment pipeline. It is easiest to understand the concept with help a diagram such as the one below: Figure 1. Example of a Deployment Pipeline (Vainio 2023, picture modified from Farley & Humble 2010). The deployment pipeline in Figure 1 is triggered by a change to the software’s code base and is then followed by multiple stages of testing. Finally, if tests are successful, the software should reach a releasable state. Given that the team is already working with a deployment pipeline, it presents an opportunity to integrate automated security tests into this process. Integrating Security Since Continuous Delivery aims for software that is always releasable, this means that the security posture of the software and the related deployment infrastructure must also be in a secure, releasable state. It might seem obvious by now, but the below diagram shows how security tests can be bolted onto the deployment pipeline: Figure 2. Example of an Enhanced Deployment Pipeline (Vainio 2023). As seen in Figure 2 above, security tests fit right into the deployment pipeline. Ideally, the security tests are run in parallel to the existing tests. This is the desired implementation that aims for security tests that don’t slow down the pipeline execution and thus the delivery process. It seems obvious that security should be one of the characteristics of the software that is tested during the deployment pipeline. But as often happens, this simple idea can be tricky to implement in practice unless you are a security expert, and that’s why I wanted to study and discover the practical ways for anyone working on the delivery process to find effective ways to integrate security tests into the process. My Master’s Thesis describes a security framework based on these core ideas. Following the thesis, in my company we have internalized this approach and developed additional practical examples and information around the security tools and practices. It is still very early days for the full-scale adoption of the framework, but we have started the journey to fully embrace the idea that security has to be an integral part of everything that we deliver. References Farley, D. & Humble, J. (2010). Continuous Delivery. Reliable Software Releases Through Build, Test and Deployment Automation. Boston: Pearson Education, Inc Vainio, M. (2023). Practical Framework for Continuous Delivery: Master´s Thesis. Metropolia UAS. 77 pages. https://www.theseus.fi/handle/10024/810697 About the author Mike Vainio is a double alumnus of Metropolia University of Applied Sciences. He first graduated as an Engineer of Information and Communication technology (Bachelor, 2014) and then graduated as a Master in Business Informatics in December 2023. Among other professional topics, Mike has a keen interest in security in software development.
A More Strategic Approach to Managing a Company’s Patent Portfolio
Half of the economic growth in industrial countries derives from innovations (Gassmann 2021: 4, 8). As a result, understanding of the significance of intellectual property (IP) rights and the value of patents has been growing during the past decades. Patents come into this picture since it is essential to protect innovations. However, protecting innovations and capturing the underlying value require a more strategic approach to Patent portfolio management. Traditionally, patents have been seen as quite a costly investment, especially for small and midsized companies. But even if many companies invest in patenting, they are not always focusing on managing their patent portfolios strategically. Yet, if managed well, the value produced by patents will probably exceed the costs in the long run and will also prevent financial losses due to lack of IP protection. Strategic Patent Portfolio Management Requires Commitment on All Organizational Levels Patenting strategy should be clear and aligned with the general company strategy. Even more importantly, it should be implemented consistently within a company. This requires cooperation and commitment from different stakeholders, from multiple organizational levels. Here, the company culture and engagement of top management have vital impact. (Jell 2012: 115; Agostini et al 2023: 1055-1056.) When Patent portfolio management is run well, patents can bring such advantages as: Bringing in market revenues (e.g. from product-related patents) Blocking competitors from using the patented technology, which creates comparative competitive advantage Increasing the attractiveness of products in marketing Enhancing the company reputation Bringing in direct licensing revenue Opening access to patent pools and cross-licensing (Gassmann 2021: 10, Jell 2012: 11). To capture the value of patents, different stakeholders, such as technical experts, product owners, business unit decision makers, and patent experts must collaboratively find answers to the questions like: Why is the patent sought? What are costs vs. benefits? How does the patent affect company´s main markets and competitor operations? Intellectual Property Management Systems Facilitate Patent Management Modern Intellectual Property Management System (IPMS) software offers multiple ways for facilitating patent management and make it more interactive for stakeholders. In practice, all the materials, documents, discussions, and decision can be stored within the IPMS in an easy and secure way. IPMS can also be used to keep all materials and decisions, and also the tacit knowledge behind the decisions available for all stakeholders and thus, making the patenting process smoother and more transparent. At the same time, utilizing IPMS as a tool efficiently requires commitment from the stakeholders. Process Development for a More Strategic Patent Management The objective of my Master´s thesis was to improve the patenting process in the case company and increase the commitment of stakeholders by developing a consistent patenting process model. The thesis also discussed utilizing the advantages of a modern IPMS data platform in the context of a defensive patent strategy. The development proposal emphasized the significance of stakeholder training and engagement. The process model which was developed as the outcome of my Master´s thesis offers a structured way to manage patenting process and gather important aspects to consider in each process phase. It also offers a model for evaluating patents during the patenting process, which was developed to meet the needs of the case company. Based on the thesis investigation, the data from each process phase was recommended to be gathered in the IPMS systematically, by utilizing visual tools which are available in the IPMS. The case company approved all the outcomes for implementation and started the change process towards new IPMS already during the final stages of the Master´s thesis. The full-scale implementation will follow early in 2024. The development proposal from the thesis helped the case company to make important commitment to the implementation of a new, more strategic and consistent patent management process. About the author Jaana Huusko is the alumnus of Master´s degree program in Business Informatics. Jaana is keen on managing patents and wants to bring a strategic touch to this exciting professional area. References Agostini, L., Nosella, A., & Holgersson, M. (2023). Patent management: The prominent role of strategy and organization. European journal of innovation management, 26(4), 1054-1070. Available from: https://doi.org/10.1108/EJIM-09-2021-0452 (Accessed 21 September 2023) Gassmann. (2021). Patent Management. Springer International Publishing (e-book). Available from: https://link-springer-com.ezproxy.metropolia.fi/book/10.1007/978-3-030-59009-3 (Accessed 18 February 2023) Jell, F. (2012). Patent Filing Strategies and Patent Management. Gabler Verlag. (e-book) Available from: https://link-springer-com.ezproxy.metropolia.fi/book/10.1007/978-3-8349-7118-0#toc (Accessed 18 February 2023) Huusko, J. 2023 Management and Evaluation of Patent Portfolio in the Context of Defensive Patent Strategy – Development of a Consistent Model for Collecting and Analysing Data Using a Modern Data Platform as a Tool. Metropolia UAS, Master´s thesis, 94 pages. Available from: https://urn.fi/URN:NBN:fi:amk-2023112231005