With the holiday season, everyone is making their wish list and checking it twice. This year, for the first time in many years, “lists” have new importance since my granddaughter just turned 5 – fully immersed in the holiday spirit – which includes her list and asking what is on Poppop’s list. Her brother, 3 ½, has a grasp on the holidays, sings along with seasonal songs, and loves lights on houses. In chatting about this with our marketing folks, they asked me to jot down a list for directors of supply chain based on my extended time in the trenches. Observe I have used the term “extended” – today’s supply chain management has more depth and breadth than 25 years ago when the concept and term first emerged to manage the end-to-end demand-supply network to avoid “on a hunch start a bunch”.
- Just in time with great fanfare in the 1990s; if only “just in time” actually worked “all of the time”. Simply synchronize your demand-supply network (DSN) such that the component part needed arrives just in the “Nick” of time. When it does not, a critical bottleneck resource goes idle and that capacity can never be recovered. Alas, a corollary is it would be great if the supply chain management firm you hired had a person who understood manufacturing and had read Factory Physics.
- Zero manufacturing excursions; if only manufacturing operated according to the planning parameters in your supply planning or matching assets with demand (MAD) model. All parameters (capacity available, lead time, yield, etc.) in your matching model are estimates that are treated as fixed across time. However, negative events (interruptions in production, the arrival of component parts, and distribution) are either yes or no. Your planning parameter may have tool availability as 90%, but on any given day the tool is working or not working. This creates challenges in planning to protect against incursions, and the need for rapid response – driving complexity (that lean had wished away).
- The appearance of demand for finished goods inventory has been slow-moving, but you have kept on the books at full price. When your firm made the product, you were sure it would sell (or at least executed to this belief). Now you have a product in inventory that has not moved. What to do? Hope the Grinch goes on a buying spree.
- Hiring “Merlin the Wizard” to lead your sales prediction team and compensate for limitations of last year’s machine learning (ML) and data science effort. The holy grail of demand management is a better ability to estimate the future. One can often read that a 10% improvement in forecasting can improve organizational performance by 50%. At the same time, the ML and Big Data Analytics folks have run a bit of python code and demonstrated a 20% improvement. Alas, your DM team jumped in and hired big data experts who have been whipping through one ML method after another. Now the folks funding the tasking are asking to see the improvement. If you only had read “Why Most Big Data Analytics Projects Fail” last year
- While Merlin is here, reduce demand variability.
- Your best planners and schedulers are not on a vacation disconnected from the internet when an emergency appears. An early lesson I learned was the purpose of a plan was to position the firm to respond to emerging situations. Think about the deployment of fire and emergency medical personnel. Alas if your best “carbon-based life forms” are on vacation, this critical component to effective response is missing – Silicon by itself has limitations.
- Limiting the impact of procurement, finance, and IT on your decisions to enhance your SCM software to enhance responsiveness and drive community intelligence – yes everyone has an opinion on SCM, most often wrong. In many organizations, SCM is viewed as “easier” than areas such as manufacturing engineering or sales or maintaining the network. Hence others often have an opinion without much analysis.
- Your optimization model to balance supply and demand does not take an unexpected excursion into the twilight zone just as the fall planning cycle begins. Although optimization models (mixed integer programming primarily) to match assets with demand or balance these are extremely helpful when the demand-supply network has alternatives, the underlying “solvers” are notorious for being “sensitive” to what appear to be unimportant changes in the values passed into the model. Although optimization software and experts are getting ever better to keep the “solver” running steady – occasional hiccups happen.
- A quiet year for outside factors (trade negotiations, weather, strikes, military events, pandemics, etc.) that impact organizational performance.
- Your organization recognizes the importance of supply chain management (SCM) and promotes you from director to senior vice-president.
All of Arkieva wishes you the best during this holiday season and looks forward to collaborating with you in 2021 to tackle the “ongoing challenge” of improving the performance of your extended supply chain.
Remember complexity exists whether you prepare for it or not, best to prepare for it.