What should we do if the customer's order quantity doesn't meet the MOQ (Minimum Order Quantity)?

In procurement and supply chain management, many companies encounter a myriad of challenges, not just on the supply side, but more significantly on the demand side. Over the years, there has been a significant differentiation in demand, with many industries experiencing smaller batch sizes and faster demand changes. Consequently, the accuracy of forecasts has become a critical issue in the supply chain. Whenever I discuss these issues in training sessions, many participants hope that forecasts could be more precise, much like a line from "My Uncle Yules": "What a surprise it would be if Yules were on this boat!" What I'd like to emphasize is that even if Yules were on that boat, the accuracy of forecasts wouldn't significantly improve. Therefore, you shouldn't rely on forecast accuracy.
The accuracy of forecasts can be enhanced, but there comes a point where further improvements become increasingly limited. It's akin to baseball; once a batting average surpasses 30%, it signifies a skilled professional player. Beyond that point, achieving higher averages becomes extremely challenging. Throughout history, players with a batting average above 40% can be counted on one hand. Even with a 40% average, a significant number of balls are not hit. Forecasting involves predicting the future, which, much like a swiftly thrown baseball, is fraught with uncertainty. As an English saying goes, "nobody has a crystal ball," meaning no one can see the future clearly. As the old adage goes, "If we knew what would happen three days in advance, we could all be rich and prosperous for a thousand years." If three days' predictions are so uncertain, what about three weeks or three months? The simple truth is that you don't know what you don't know. Once you've exhausted what you know or might know, what remains is what you don't know you don't know. At this point, you either leave it to chance or seek an alternative. Stubbornly fixating on forecast accuracy is akin to repeatedly banging your head against a wall—it might seem like effort, but there's little reward in reality.

Before seeking an alternative, let's ponder a question: Why forecast? Some might say it's due to the uncertainty of demand. However, demand uncertainty is a manifestation, not the essence. For instance, consider tap water; sometimes your household uses more, sometimes less. The demand is indeed uncertain, yet you never forecast your own usage—turn the tap, and water flows instantly, with a response time of zero. So, why bother with forecasting? It's because there's a response time, which involves product manufacturing, transportation, and delivery. Why not produce products in advance and distribute them to consumption points, ensuring that when demand arises, they're readily available, much like tap water? Herein lies the issue of inventory costs and risks. Understanding these points reveals that finding an "alternative" means finding ways to shorten the response cycle, including production, transportation, and delivery times. If that isn't feasible, then strategies to mitigate inventory risks by locating inventory closer to consumption become crucial. This all falls within the domain of supply chain operations. In essence, supply chain operations are part of the solution to achieving forecast accuracy.

Firstly, shortening the response cycle involves reducing production, transportation, delivery, and waiting times for products. Shortening this cycle essentially means accelerating both product and information flows. Lean manufacturing, a widely practiced approach for years, aims to hasten product flows. For example, adopting one-minute mold changes reduces setup times, while employing preventive maintenance reduces equipment downtime. Creating manufacturing centers minimizes wait times during production processes. These efforts, in various ways, facilitate faster product flow, thereby shortening the production cycle. Implementing 5S practices on the production floor—keeping tools and materials in convenient locations—ensures smoother operations and more effective product flows. Improving suppliers' on-time delivery rates and ensuring product quality likewise reduce downtime during production and the time spent handling quality issues, ensuring the smooth flow of products. Across several decades and industries, efforts to expedite product flows have focused on enhancing processing, transportation, and distribution, aiming for speed, lower costs, and better quality.

However, within the product's response cycle, only a fraction involves actual movement. Most of the time, products remain stagnant as companies navigate various processes, namely, information flows. Despite efforts to shorten the production cycle, the effectiveness of lean manufacturing is limited because the time taken by information flows remains unchanged. For instance, in some companies, customer orders undergo a 1 to 3-day contract review to ensure understanding; approval before entering an ERP takes another 1 to 3 days to ensure correct decisions. It takes another day to input into the ERP, and then another day to run the MRP, finally generating orders for suppliers. Before sending the orders to suppliers, they undergo additional approvals based on the order's value, which again takes 1 to 3 days. Once approved, orders are placed on electronic trading platforms or emailed to suppliers, taking an additional day. Throughout this process, product flow is stagnant while only information flows. In certain companies, it takes 2 to 3 weeks for demand information to reach the primary supplier. As this information cascades down the chain, reaching third or fourth-tier suppliers, five to six weeks elapse. This explains why delivery cycles for some products span from three to six months. Longer delivery cycles increase dependence on forecasting.

Inefficient information flows pose a significant problem. Recognizing this, considerable efforts have been made to accelerate information flows, such as adopting information technology and e-commerce to embed certain decisions into systems. Flattening organizational structures to decentralize decision-making and shorten decision cycles has been another strategy. However, compared to efforts in accelerating the production cycle, improvements in information flow are still insufficient. This presents an opportunity for reducing response cycles. A holistic improvement in end-to-end supply chain processes focuses primarily on enhancing information flows, with supply chain operations playing a pivotal role.

The response cycle for products cannot infinitely shorten due to physical laws. For instance, irrespective of the transportation speed used, it still takes at least 10 hours to travel from North America to Asia because you cannot defy physical laws by instantly transporting goods, similar to sending an email. No matter how lean or well-planned the production capacity is, machining products using lathes, mills, or grinders takes time. Therefore, there's a limit to shortening response cycles; going beyond that threshold becomes prohibitively expensive. This necessitates addressing inventory aspects: can we push inventory directly to where consumption occurs, much like how tap water is delivered to your kitchen? This way, when demand arises, products are readily available.

However, inventory carries risks; prolonged inventory might depreciate, and surplus stock might need to be written off if demand diminishes. How do we then mitigate inventory risks? This is where standardized design comes into play.

Standardizing design operates on multiple levels: (1) Utilizing industrial-standard components, which is the optimal form of standardization; (2) Even for customized items, using the same customized part across multiple products constitutes a form of standardization; (3) While the product design may not be standardized, standardizing the manufacturing process reduces production complexity and shortens production cycles. Standardization benefits from economies of scale and risk pooling. The advantages of scale are evident. Risk pooling is straightforward: the more standardized the product, the lower the risk of inventory pile-up because if one customer doesn't need it, another might, thereby reducing the demand for precise forecasts. Similarly, with more customers using standardized