Data Analytics in Supply Chain Decisions

Choosing the right chart is half the battle in data storytelling. This one visual helped me go from “𝐖𝐡𝐢𝐜𝐡 𝐜𝐡𝐚𝐫𝐭 𝐝𝐨 𝐈 𝐮𝐬𝐞?” → “𝐆𝐨𝐭 𝐢𝐭 𝐢𝐧 10 𝐬𝐞𝐜𝐨𝐧𝐝𝐬.”👇 𝐇𝐞𝐫𝐞’𝐬 𝐚 𝐪𝐮𝐢𝐜𝐤 𝐛𝐫𝐞𝐚𝐤𝐝𝐨𝐰𝐧 𝐨𝐟 𝐡𝐨𝐰 𝐭𝐨 𝐜𝐡𝐨𝐨𝐬𝐞 𝐭𝐡𝐞 𝐫𝐢𝐠𝐡𝐭 𝐜𝐡𝐚𝐫𝐭 𝐛𝐚𝐬𝐞𝐝 𝐨𝐧 𝐲𝐨𝐮𝐫 𝐝𝐚𝐭𝐚: 🔹 𝐂𝐨𝐦𝐩𝐚𝐫𝐢𝐬𝐨𝐧? • Few categories → Bar Chart • Over time → Line Chart • Multivariate → Spider Chart • Non-cyclical → Vertical Bar Chart 🔹 𝐑𝐞𝐥𝐚𝐭𝐢𝐨𝐧𝐬𝐡𝐢𝐩? • 2 variables → Scatterplot • 3+ variables → Bubble Chart 🔹 𝐃𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧? • Single variable → Histogram • Many points → Line Histogram • 2 variables → Violin Plot 🔹 𝐂𝐨𝐦𝐩𝐨𝐬𝐢𝐭𝐢𝐨𝐧? • Show part of a total → Pie Chart / Tree Map • Over time → Stacked Bar / Area Chart • Add/Subtract → Waterfall Chart 𝐐𝐮𝐢𝐜𝐤 𝐓𝐢𝐩𝐬: • Don’t overload charts; less is more. • Always label axes clearly. • Use color intentionally, not decoratively. • 𝐀𝐬𝐤: What insight should this chart unlock in 5 seconds or less? 𝐑𝐞𝐦𝐞𝐦𝐛𝐞𝐫: • Charts don’t just show data, they tell a story • In storytelling, clarity beats complexity • Don’t aim to impress with fancy visuals, aim to express the insight simply, that’s where the real impact is 💡 ♻️ Save it for later or share it with someone who might find it helpful! 𝐏.𝐒. I share job search tips and insights on data analytics & data science in my free newsletter. Join 14,000+ readers here → https://lnkd.in/dUfe4Ac6

One of the biggest challenges in data visualization is deciding 𝘸𝘩𝘪𝘤𝘩 chart to use for your data. Here’s a breakdown to guide you through choosing the perfect chart to fit your data’s story: 🟦 𝗖𝗼𝗺𝗽𝗮𝗿𝗶𝘀𝗼𝗻 𝗖𝗵𝗮𝗿𝘁𝘀 If you’re comparing different categories, consider these options: - Embedded Charts – Ideal for comparing across 𝘮𝘢𝘯𝘺 𝘤𝘢𝘵𝘦𝘨𝘰𝘳𝘪𝘦𝘴, giving you a comprehensive view of your data. - Bar Charts – Best for fewer categories where you want a clear, side-by-side comparison. - Spider Charts – Great for showing multivariate data across a few categories; perfect for visualizing strengths and weaknesses in radar-style. 📈 𝗖𝗵𝗮𝗿𝘁𝘀 𝗳𝗼𝗿 𝗗𝗮𝘁𝗮 𝗢𝘃𝗲𝗿 𝗧𝗶𝗺𝗲 When tracking changes or trends over time, pick these charts based on your data structure: - Line Charts – Effective for showing trends across 𝘮𝘢𝘯𝘺 𝘤𝘢𝘵𝘦𝘨𝘰𝘳𝘪𝘦𝘴 over time. Line charts give a sense of continuity. - Vertical Bar Charts – Useful for tracking data over fewer categories, especially when visualizing individual data points within a time frame.    🟩 𝗥𝗲𝗹𝗮𝘁𝗶𝗼𝗻𝘀𝗵𝗶𝗽 𝗖𝗵𝗮𝗿𝘁𝘀 To reveal correlations or relationships between variables: - Scatterplot – Best for displaying the relationship between 𝘵𝘸𝘰 𝘷𝘢𝘳𝘪𝘢𝘣𝘭𝘦𝘴. Perfect for exploring potential patterns and correlations. - Bubble Chart – A go-to choice for three or more variables, giving you an extra dimension for analysis. 🟨 𝗗𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻 𝗖𝗵𝗮𝗿𝘁𝘀 Understanding data distribution is essential for statistical analysis. Use these to visualize distribution effectively: - Histogram – Best for a 𝘴𝘪𝘯𝘨𝘭𝘦 𝘷𝘢𝘳𝘪𝘢𝘣𝘭𝘦 with a few data points, ideal for showing the frequency distribution within a dataset. - Line Histogram – Works well when there are many data points to assess distribution over a range. - Scatterplot – Can also illustrate distribution across two variables, especially for seeing clusters or outliers. 🟪 𝗖𝗼𝗺𝗽𝗼𝘀𝗶𝘁𝗶𝗼𝗻 𝗖𝗵𝗮𝗿𝘁𝘀 Show parts of a whole and breakdowns with these: - Tree Map – Ideal for illustrating hierarchical structures or showing the composition of categories as part of a total. - Waterfall Chart – Perfect for showing how individual elements contribute to a cumulative total, with additions and subtractions clearly represented. - Pie Chart – Suitable when you need to show a single share of the total; use sparingly for clarity. - Stacked Bar Chart & Area Chart – Both work well for visualizing composition over time, whether you’re tracking a few or many periods. 💡 Key Takeaways - Comparing across categories? Go for bar charts, embedded charts, or spider charts. - Tracking trends over time? Line or bar charts help capture time-based patterns. - Revealing relationships? Scatter and bubble charts make variable correlations clear. - Exploring distribution? Histograms or scatter plots can showcase data spread. - Showing composition? Use tree maps, waterfall charts, or pie charts for parts of a whole.

As supply chain managers look towards manufacturing industry verticals that may see volume growth in 2024, in addition to looking at trends in industrial production indexes (which measure physical unit output), it is also useful to look at the Federal Reserve Board’s capacity utilization indexes for these industries. The reason is simple: an industry that is operating at a high level of capacity utilization likely has little upward room for output growth given strains placed on equipment and labor. To illustrate, below are two capacity utilization indexes from the FRB. Thoughts: •The top index shows seasonally adjusted capacity utilization for plastics & rubber products manufacturing (https://lnkd.in/gPYc_GMb). Capacity utilization has fallen very sharply starting in Q4 2022 and has yet to recover. Looking at year-over-year industrial production (https://lnkd.in/gB3PaQKu), the decline has been about 5%. Thus, if conditions improve, we may could see an increase in output of ~5% as an upper bound (best-case scenario). •The bottom index shows capacity utilization for nonmetallic mineral product manufacturing (https://lnkd.in/gYe4piEh). Capacity utilization since early 2022 has been running 8-10 percentage points above 2018 and 2019 levels, suggesting a sector where we have little room for additional growth in output unless additional capacity is added. Thus, while industrial production has remained strong (https://lnkd.in/gtgD-YM4), I see fewer opportunities for output volume to grow in 2024 even if demand manifests (since these factories are likely to begin to encounter supply constraints). •For anyone interested, here is a link (https://lnkd.in/gy2ukdvr) to all the capacity utilization indexes the FRB publishes. Implication: augmenting industrial production data measuring output with capacity utilization data can provide a more comprehensive picture of how a manufacturing sector is performing and its potential to increase output if demand conditions were to improve. More free competitive intelligence data to incorporate into strategic decision making (whether as a trucking manager or sourcing professional trying to anticipate supplier lead times). #supplychain #supplychainmanagement #manufacturing #economics #freight #trucking

Supply chain planning cannot breathe without metrics. This infographic shows 7 critical metrics: # 1 - WMAPE (Weighted Mean Absolute Percentage Error) ↳ (SUM(ABS(Forecast - Actual)) / SUM(Actual)) * 100 ↳ Measures forecast accuracy. Lower = better.   # 2 – Bias ↳ (SUM(Forecast - Actual) / SUM(Actual)) * 100 ↳ Shows if forecasts over- or under-estimated demand   # 3 – OTIF ↳ (SUM(Orders Delivered On Time and In Full) / SUM(Total Orders)) * 100 ↳ Service level; orders delivered as promised   # 4 - Inventory Turnover ↳ SUM(COGS) / ((Beginning Inventory + Ending Inventory) / 2) ↳ How fast inventory is sold and replaced   # 5 Plan Attainment ↳ (SUM(Actual Output) / SUM(Planned Output)) * 100 ↳ Execution vs. plan reliability   # 6 - Cash conversion cycle (CCC) ↳ CCC = DIO + DSO – DPO ↳ DIO = (Average Inventory / SUM(COGS)) * 365 ↳ DSO = (Accounts Receivable / SUM(Revenue)) * 365 ↳ DPO = (Accounts Payable / SUM(COGS)) * 365 ↳ Days to turn cash outflows into inflows   # 7 - EBITDA ↳ Net Income + SUM(Interest) + SUM(Taxes) + SUM(Depreciation) + SUM(Amortization) ↳ Profit from core operations Any others to add?

Fascinating real-world case study from Walmart on supply chain optimization. They built an approach where the decision to build a new DC and the choice to reroute a single truck are part of the same conversation. A few thoughts: 1 - Simulation will be more important than ever with all of the volatility going on. Their simulation platform runs what-if scenarios that helps leadership stress-test network changes before committing capital. 2 - The load planner function is impressive. It accounts for DOT rules, temperature zones, axle weights, and loading patterns down to the pallet. 3 - From a strategy perspective, the four-tier system: strategic network design, facility alignment/capacity planning, execution planning, and real-time/dynamic execution is really solid. Each layer pushes and pulls on the others so decisions stay relevant.

Over the last several months I’ve been thinking deeply about yard scheduling and sequencing as part of transforming Toyota North America’s supply chain and logistics operations, I’ve spent a lot of time thinking about how to bring together theory and real-world execution. Traditional optimization models can be elegant in theory (centralized, end-to-end, globally optimal) but they tend to collapse under real-world complexity. Uncertain arrivals, variable processing times, unpredictable labor shifts, and equipment issues create a level of volatility that static plans simply can’t keep up with. And while rule-based systems offer more robustness in the face of this noise, they often leave too much efficiency on the table. That’s why I’ve been drawn to the framework of Sequential Decision Analytics (SDA), developed by Warren Powell. SDA doesn’t try to force perfect optimization onto an imperfect world. Instead, it gives us a way to structure decision-making over time under uncertainty. It breaks problems into stages, accounts for new information as it arrives, and lets us build policies that adapt as the system evolves. It respects the fact that operations happen in real-time and decisions today affect what options are available tomorrow. That’s exactly the kind of thinking required in a yard environment where vehicles move through multiple stations (unloading, parking, staging, fueling, processing) and each decision has ripple effects downstream. In my proposed implementation, we use a hybrid model. A short-term plan is “frozen” to give operators clarity and confidence. Outside that window, the system uses agentic AI (intelligent agents embedded across the yard) to make real-time adjustments based on observed state. These agents use SDA principles: observing the current state, making decisions based on local policies, learning from outcomes, and aligning to overall objectives like throughput and delay reduction. The idea is to use reinforcement learning to simulate downstream consequences and constantly refine those policies. What I appreciate about SDA is that it provides a structured way to balance global coordination with local flexibility. It doesn’t assume perfect data or perfect models. It gives us a way to build intelligent systems that learn and adapt, without sacrificing stability on the ground. As supply chains get more dynamic, more interconnected, and more complex, this kind of thinking becomes essential. #SupplyChain #Optimization #RLSO #SDA #OperationsResearch #MachineLearning

Dear My Network, I'm wrapping this series on Segmentation with the following key Takeaways: • ML and Agentic AI are powerful enablers of E2E supply chain segmentation by enhancing agility, automation, and intelligence across supply chain processes. • These technologies can dynamically adapt segmentation strategies based on real-time data, customer behavior, and changing market conditions. • It can identify profitable clusters, predict disruptions, and automate scenario planning across multiple supply chain models. • Agentic AI brings autonomy to processes—executing tasks, learning, and optimizing supply chain responses without constant human intervention. The insights for 4-part series are drawn from my chapter in our new book: https://lnkd.in/gVNSdWsW Lets close with another Example: Global Consumer Electronics Manufacturer - Context: A multinational consumer electronics company sells both premium and value-tier products across multiple channels—direct-to-consumer (DTC), big-box retailers, and e-commerce platforms. Each segment had distinct demand patterns, service expectations, and profitability margins. - Challenge: They were using a one-size-fits-all supply chain model, leading to: • Stockouts of premium products during product launches • Overstocking of slower-moving value-tier items • High logistics costs due to expedited shipments - E2E Segmentation in Action: 1. Planning Phase They used ML algorithms to profile and cluster customers and products based on buying behaviors, seasonality, margin contribution, and service requirements. 2. Implementation Phase They designed virtual supply chains: • One for high-margin flagship unpredictable products with make-to-order and expedited fulfillment • Another for value-tier SKUs using a low-cost, forecast-driven model with bulk shipments • A third for e-commerce with decentralized inventory and last-mile delivery partners 3. Sustain Phase Agentic AI systems monitored these segments in real time, dynamically adjusting planning parameters and alerting teams when service levels or cost thresholds were breached. - Results: • 15% reduction in working capital tied to inventory • 10% improvement in on-time delivery for premium products • Faster decision-making and fewer fire drills • Greater alignment between sales, supply chain, and finance This example reflects the core principles outlined in my book chapter on segmentation, showing how advanced technology and structured transformation can drive real business value. Now, How are you planning to use AI to enable e2E segmentation in your supply chain? Please share your thoughts in the comments!

Forecasting solutions touting the use of AI/ML models are hard to avoid these days. But there is a hidden risk that most companies tend to ignore. The latest models are great but I worry these are being applied in a way that will only amplify “the bullwhip effect”. What do I mean? Bull whip effect is the distortion of the demand signal as it travels from the consumption end of the supply chain to the production end, while traversing multiple physical and informational nodes along the way. As a result, the demand signal at the production end could be orders of magnitude variable than actual consumption. This is nothing new and we have known about this effect for two decades plus. As we apply algorithms to forecasts, unless we account for the bullwhip effect, we are bound to amplify distortion despite best intentions. Now, this is not about outlier elimination which I believe algorithms do a pretty good job of eliminating. I am talking about misinterpreting noise as signal, over-interpreting variability and causing inventory gyrations that ultimately hurt customers. A classic example is using order/shipment data at Distribution Centers for forecasting or worse yet, factory shipments as a proxy for demand. Most S&OP plans only focus on order and shipment data without systematically factoring in channel inventory and demand. So what is the fix? In my opinion, if you are a consumer company (CPG, Hi-Tech, Retail, Pharma/Healthcare, and even manufacturing), build the capability to forecast a demand signal that is as close to the final consumption point. For example, a CPG brand could forecast retail/e-commerce sell-through demand, normalize it for channel inventory and then propagate that signal up into the supply chain. And the best part - those same AI/ML models will work much better for you. To be honest, B2B and industrial companies also benefit from a similar approach by getting closer to end customer demand. Better yet, this unlocks better demand intelligence which fuels better S&OP judgements, new product forecasting quality, lifecycle management, capacity planning and more. If you are looking for a 10x transformation, this is one of them. It’s bizarre to me when I see companies side-stepping this fundamental step and then complain about forecast accuracy, or data cleanliness or something else hurting their supply chain service levels and costs. Leaders who are pursuing unlocking growth from their supply chains while reducing cost-to-serve need to lead from the front in championing this capability. 

Over the past few weeks, I've frequently been asked by the leaders I’m advising to share key performance indicators (KPIs) that organizations should consider as they embark on their supply chain digital transformation journeys. This has sparked important conversations about the necessity of aligning metrics with specific organizational goals and execution strategies. It also opened the door for candid conversations as to the need to engage staff in the process to allow for organizational enrollment which is crucial to long term success as it fosters a shared understanding of what success will look like and helps ensure that the selected metrics are tailored to your unique business context. Given the frequency of these requests, I thought it might be helpful to many to share a few common starting points for organizations to consider with the understanding that these must be tweaked for your own journey and this list is certainly not exhaustive… Digital Adoption Rate: Track the extent to which supply chain processes have been digitized, indicating progress in transformation. Order Fulfillment Rate: Track the percentage of customer orders fulfilled on time and in full. Inventory Turnover: Measure how frequently inventory is sold and replaced, highlighting efficiency in inventory management. Supply Chain Cycle Time: Assess the total time from order initiation to fulfillment, revealing areas for improvement. Perfect Order Rate: Evaluate the percentage of orders delivered on time, complete, and undamaged. Cost to Serve: Understand the total costs associated with fulfilling customer orders, including logistics and overhead. Forecast Accuracy: Monitor how closely your demand forecasts align with actual sales to enhance planning. Return on Supply Chain Investments (ROSI): Measure the financial returns from your investments in supply chain technologies and processes. Supplier Lead Time: Analyze the average time taken by suppliers to deliver goods, impacting your operations. Customer Satisfaction Score (CSAT): Gauge how satisfied customers are with product availability and order fulfillment. Curious to know what others think of this list as well…. #SupplyChain #DigitalTransformation #KPIs #Leadership #BusinessGrowth

Tariff volatility is here. Can you adapt fast enough? Entering 2025 we are facing a radically altered trade landscape. Tariff proposals range from 10% to 60%.  🚢 Organizations must manage rising costs, sudden supply disruptions, and inflationary pressures, all while contending with fast-changing rules and potential retaliation from trading partners. Yet volatility also creates opportunities for organizations who are prepared. 🧭 𝗚𝗿𝗮𝗽𝗵-𝗯𝗮𝘀𝗲𝗱 𝗱𝗮𝘁𝗮𝗯𝗮𝘀𝗲𝘀 𝗮𝗻𝗱 𝗮𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 𝗰𝗮𝗻 𝗽𝗿𝗼𝘃𝗶𝗱𝗲 𝗿𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗶𝗻𝘀𝗶𝗴𝗵𝘁𝘀 𝗶𝗻𝘁𝗼 𝘆𝗼𝘂𝗿 𝗶𝗻𝘁𝗲𝗿𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗲𝗱 𝘄𝗲𝗯 𝗼𝗳 𝘀𝘂𝗽𝗽𝗹𝗶𝗲𝗿𝘀, 𝘁𝗮𝗿𝗶𝗳𝗳𝘀, 𝗮𝗻𝗱 𝗹𝗼𝗴𝗶𝘀𝘁𝗶𝗰𝗮𝗹 𝗿𝗼𝘂𝘁𝗲𝘀. Here's how: 1️⃣ 𝗠𝘂𝗹𝘁𝗶-𝗛𝗼𝗽 𝗦𝘂𝗽𝗽𝗹𝘆 𝗖𝗵𝗮𝗶𝗻 𝗩𝗶𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝘆 ↳ Map your entire supplier network as nodes and relationships in a graph.  ↳ Visualize dependencies several layers deep, often hidden in traditional systems. 2️⃣ 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗧𝗮𝗿𝗶𝗳𝗳 𝗦𝗰𝗲𝗻𝗮𝗿𝗶𝗼 𝗠𝗼𝗱𝗲𝗹𝗶𝗻𝗴 ↳ Add tariffs to the graph and then use graph algorithms to simulate alternate sourcing paths with lower duties or better resilience. ↳ This enables decision-makers to test “what-if” scenarios, minimizing guesswork when a sudden tariff spike occurs. 3️⃣ 𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝘃𝗲 𝗥𝗶𝘀𝗸 & 𝗗𝗲𝗽𝗲𝗻𝗱𝗲𝗻𝗰𝘆 𝗔𝗻𝗮𝗹𝘆𝘀𝗶𝘀 ↳  Apply centrality and community-detection algorithms to find which suppliers or markets could cause cascading failures. ↳  Uncover clusters of high-risk exposure, allowing proactive adjustments rather than reactive damage control. Graph-based platforms help executives move beyond spreadsheets and siloed databases. They offer a living, interconnected view of all the moving parts, enabling better-informed decisions on pricing, sourcing, and expansion. 🚀 𝗔𝘁 𝗗𝗮𝘁𝗮2 𝘄𝗲 𝗵𝗮𝘃𝗲 𝗯𝘂𝗶𝗹𝘁 𝗼𝘂𝗿 𝗿𝗲𝗩𝗶𝗲𝘄 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗼𝗻 𝘁𝗼𝗽 𝗼𝗳 𝗡𝗲𝗼4𝗷 𝘁𝗼 𝗵𝗲𝗹𝗽 𝗼𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝘀 𝗮𝗰𝗰𝗲𝗹𝗲𝗿𝗮𝘁𝗲 𝘁𝗵𝗲𝗶𝗿 𝗮𝗱𝗼𝗽𝘁𝗶𝗼𝗻 𝗼𝗳 𝗴𝗿𝗮𝗽𝗵𝘀 𝗮𝗻𝗱 𝗿𝗲𝗹𝗶𝗮𝗯𝗹𝗲 𝗔𝗜 𝗳𝗼𝗿 𝗰𝗿𝗶𝘁𝗶𝗰𝗮𝗹 𝗮𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀. If your organization is concerned about how it can adapt to the new era of trade volatility, reach out and we can start the conversation. ♻️ Know someone who needs better visibility into their supply chain? Share this post to help them out! 🔔 Follow me Daniel Bukowski for daily insights about delivering value from connected data.