Approaches To Maintain Practicality In Innovative Engineering Solutions

I often discuss strategy for public adoption of new technologies -- like autonomous aircraft and advanced air mobility -- as well as creative innovation strategies, whether for incremental, disruptive, and breakthrough innovation. These discussions are usually esoteric in nature, to encourage critical thought and mindset shifts. However, this time I'd like to offer something more concrete: a simple, practical strategy to address both adoption and innovation simultaneously. First, let's talk about first and second-order effects. When introducing something new, first-order effects are the immediate and direct results, while second-order effects are the indirect consequences that occur over time and are not immediately visible. These second-order effects, or "frictions," can significantly impact the success and sustainability of a technology (or project/product/policy, etc.). For example, New York City's rent control policy aimed to provide affordable housing for veterans but led to a decline in property quality and supply, creating unintended negative outcomes. Similarly, the introduction of smartphones increased productivity but also led to burnout due to the "always-on" culture. The silver-lining is that second-order frictions often reveal unexpected problems needing solutions, opening space for innovation, which I call the "third space." The third space is where innovative solutions emerge, creating new value and accelerating the adoption of first-order technologies. These "third space accelerators" make people more comfortable adopting the original technology. This approach as strategy involves: 1. Identifying likely second-order effects early. 2. Proactively developing third space solutions. 3. Presenting the complete ecosystem to drive adoption. 4. Using the existence of solutions to comfort potential adopters. Third space accelerators are strategic tools. By proactively developing third space solutions, we can better mitigate the negative impacts of second-order frictions and create innovative opportunities that drive adoption and innovation simultaneously. With the example of smart phones, third space solutions like digital wellness apps and wellness integration into workspaces emerged to address these new challenges, but only after the fact. I encourage you to apply this strategy in your projects: identify potential second-order effects early, develop solutions to address them, and present a comprehensive ecosystem that includes both the problem and the solution. This approach will not only facilitate smoother adoption but also foster continuous innovation. #innovation #leadership #technology #influence

Sometimes I’m jealous of academics and their clean, tidy toy problems… But here’s how we can make their theories work in the messy real world. Decision science often involves translating theoretical concepts into actionable real-world strategies. This translation is crucial in environments where uncertainty and variability are the norms, such as in our Toyota supply chain management. Consider the theoretical frameworks that emphasize reinforcement learning and stochastic optimization. These theories provide strategies for adapting decisions dynamically as new information becomes available, similar to how a GPS recalculates routes in real-time based on traffic changes. 🎯 Practical Advice: 1️⃣ Start Small. Implement theoretical models on a small scale before rolling them out across the organization. This allows you to observe the model’s performance and make necessary adjustments. 2️⃣ Use Hybrid Models. Combine theoretical models with heuristic approaches. This can provide a balance between optimal and practical solutions, especially in complex and uncertain environments. 3️⃣ Frequent Re-evaluation. The real world is dynamic. Regularly revisit and update your models to align with new data and changing conditions. 4️⃣ Cross-functional Teams. Engage experts from various domains (data science, operations, IT) in the implementation process. Their diverse perspectives can help identify and mitigate practical challenges early. For instance, global supply chain disruptions challenge us to go beyond traditional models. Theoretical optimization might dictate certain stock levels and operational efficiencies, but real-world scenarios require us to adapt to unforeseen shortages and demand surges. The art lies in applying these adaptive, learning-based theories to continuously refine our strategies, ensuring they remain robust amidst volatility. The beauty of this approach is in its adaptability. It’s about learning from the environment and iteratively improving processes, mirroring the way algorithms learn and optimize based on new data. 💭 How do you balance the elegance of theory with the messy realities of practice in your field? #DataScience #Optimization #StochasticOptimization #ReinforcementLearning #SupplyChainManagement #OperationsResearch

The most frustrating thing about IT, Software Engineering (SWE), and Software Development (SD) professionals entering the industrial automation space is that a significant portion genuinely believes: More code equates to better functionality. Increased complexity in logic equals impressive results. Clever use cases are inherently valuable, even without practical relevance. Modern engineering schools typically emphasize languages such as C, C++, Python, and Java, often overlooking ladder logic unless it's part of a hands-on practical engineering curriculum. Even then, the teachings frequently come from Ph.D. professors with minimal real-world automation experience, leading students to adopt impractical methods like overusing alias tags or unnecessarily complex sequencers. While C is valuable for networking and embedded systems programming, higher-level languages like Python or Java rarely suit real-time automation (RTA) scenarios effectively. Despite this, many recent graduates, particularly those holding Engineering Management degrees, end up supervising PLC programming teams within industrial automation integrators. Consequently, they lean heavily on practices learned in academia: FPGA programming in Verilog, microcontroller programming in C, or applying Python-based machine vision AI solutions—rather than the practical, robust approaches needed in automation environments. So, if this message feels like it "hits", remember: As you enter or progress into our realm of industrial automation, always adhere to the fundamental principle: Keep It Simple, Stupid (KISS). Complexity and excessive cleverness rarely impress anyone who genuinely understands industrial automation. No one is saying don't add good practice or software management methods— but remember the customer is the end user: NOT your shops need for modularity and massive code libraries. You're code needs to be safe, easily read, execute and error handle well, and be easily expandable before anything. Practical, straightforward, and maintainable solutions always hold greater value.