Supply Chain Management: what is it? - PART 2

This second posting on our new blog continues the description of where SCM stands now. It is necessary in my opinion to write these postings on the status quo before I can start with the forward-looking articles that this blog is really about. This particular posting is part 2 of the article Supply Chain Management: what is it?

In the first part I delineated SCM within the enterprise application landscape at a macro level. I gave a broad definition of SCM and what is in and out of its scope in general terms, and how it is related to other enterprise domains such as ERP, CRM and MES. In this second part I intend to be more specific as to which known functional areas fall under the SCM umbrella. A warning to the casual reader: this will unfortunately involve quite a number of acronyms again.

Let’s make a short journey across the supply chain starting at the demand side and work our way towards to supply side. Along the way I’ll discuss the functional areas that exist at each step.

Demand

Undeniably demand is one of the most important, if not the most important, aspects of any supply chain. It certainly is the most important contributor to the livelihood of any production business. The roles SCM plays on demand are somewhat limited, hence demand may not be the most important focal point for any SCM implementation. For example, demand generation is owned by sales and marketing departments, and these seldom use SCM systems. Demand generation is a very involved process of reaching out to customers, prospects and other influencers in which relations are acquired, nurtured and demand is created. In contrast, SCM limits itself to the relatively laid back process of forecasting demand and at most influencing it. The first demand centric applications in the space now known as SCM were simple sales forecasting tools. At first these were limited to taking last year’s sales numbers and assuming this year’s numbers would be the same. Then they evolved into applications that could use any number of statistical and regressional methods to forecast future sales based on many years of historic data, using the best combination of methods for each individual product forecast. By the time more sophisticated functionality such as pyramid aggregation and distribution (or proration) were added these systems were commonly known as Demand Planning (DP) applications. These systems also typically output a safety stock to cover any underestimation in sales forecast and prevent stock outs. The next major improvement involves user collaboration. The impact of having the right experts enter their adjustments to a statistically calculated forecast is usually many times higher than the few extra percentage points that may be squeezed out of an already decent statistical forecast by improving on the methods. The right experts are however in the sales and marketing departments of your company and any number of people in your downstream supply chain partners’ companies. This means giving the option and the incentives for those experts to actually enter their knowledge on a continuous basis. The applications that provide that option are known as Demand Management (DM) systems. Frequently, these systems also cover the Promotions Management and Pricing & Revenue Management type functionality (thankfully, these areas do not have generally accepted acronyms of their own!). This is where SCM becomes more proactive on demand to actually influence it in order to increase revenue, rather than just forecast what it will be. For many industries this has a HUGE potential impact on the bottom line.

Once demand has been projected into the future the generated numbers can be used by many business processes. As a rule of thumb, all forward-looking uses are the domain of SCM. This can be at various levels of planning ranging from the very broad and high level Strategic Network Optimization down to operational point solutions such as Available to Promise (ATP). The high level processes will be discussed in more detail later in this article. ATP is where a sales rep (or a customer via a sales portal) asks the system whether an order can be delivered on time and in full if ordered now. The sales enquiry is checked against unconsumed forecast for that customer or group of unnamed customers. If the unconsumed quantity is larger than the requested quantity the product is available to order. This relatively simple process has more advanced counterparts and more simplified counterparts. Usually ERP systems provide an altered version where the sales enquiry is not compared against unconsumed forecast, but rather against unreserved inventory that is already on hand. The downside is that ATP in that form cannot look very far forward, which makes sense since ERP itself is not forward looking, and simply misses the required data to do the longer range version.

Note: One software application dealing directly with demand that is commonly mistaken to be part of SCM is Sales Force Automation (SFA). The only reason I mention this here is because this mistake is so common. SFA truly belongs under the CRM umbrella. It deals with relationships, not product flow.

Sourcing & Distribution

Demand trickles (or in good times, gushes) along the supply chain where one of the first stops is the distribution network. Demand from many customer locations is grouped into distribution centers and warehouses that supply those locations. This can be a multiple step distribution network where product may come from any number of plants to go to any number of distribution centers, warehouses, forward warehouses and satellite warehouses, and may in some circumstances allow inter-warehouse transfers. At each distribution step  inventory may be kept of any product that flows through that warehouse, causing a decoupling of the demand patterns accumulated on the shipping end and the required supply patterns on the receiving end of each, and an inventory management requirement in the middle. Both of these cause issues that may need to be addressed. The pattern mismatches between supply and demand are the main cause for the so-called bullwhip effect (an article on this will surely be posted on this blog in the near future, wink to my wife). This negative effect is amplified when any or all of the demand, supply or inventory management is done by different parties and information not shared amongst them. A good collaborative system that allows every party to make their plans, even fuzzy far-future plans, known without immediate contractual lock-in can adequately address this. For the inventory management requirement many applications exist in the SCM space that address it ranging from strategic via tactical down to operational. The strategic and tactical ones are typically part of a larger solution that will be covered further down in this posting. On the operational side of the range the most notable functional area is Warehouse Management Systems (WMS). These frequently extend into executional detail, making it unclear whether WMS would fall under SCM or ERP. In my opinion, the more proactive, forward-looking implementations should fall under SCM, the more reactive, present-looking implementations would fall under ERP. Which type you implement is largely dependent on the industry you are in. Where distribution concerns itself with what goes where and how long to keep it on inventory, sourcing deals with where the product comes from. The two are typically optimized hand-in-hand.

Transportation

Once you know what product needs to supply which warehouse by when and where to source it from, you can plan how to get it there. The transportation functionality, like all other functionalities in SCM, is historically segmented, but combined solutions exist where the value of the combined solution is higher than that of the sum of the parts. The two well known areas are route optimization and load planning. The former tries to solve the infamous traveling salesman problem which answers the question what is the optimal route to deliver to all locations with the shortest total miles. A modern day version may optimize for the lowest total use of fossil fuels and emissions. The latter determines which orders to place where on a truck to ensure the unloading at each location is done in an efficient manner while keeping the load balanced for the next segment of the route. In certain industries product is shipped in bulk and trucks, wagons, or compartments within them can have complex rules on what product may use it after another one, even multiple loads later. Other areas of transportation planning (TP) involve minimization of total cost by optimizing on freight rates which fluctuate highly by time of the year, area, mode of transport, and transport company. On a tactical level transportation can be one of many influences that determine an overall optimal plan using a single solution.

Manufacturing

Manufacturing is for many industries undoubtetly the area of SCM with the most complexity. And as an extension of this, the area with the most opportunity for improvement in terms of cost reduction, and revenue, cycle time and service level increase. Those industries where this statement is untrue are typically well served with the basic SCM-type functionality available in most ERP systems. The industries for which the statement is true, are the prime reason SCM even exists as its own domain. Assuming either a strategic or tactical plan has already determined how much product to manufacture at each plant, or that there is only one option (single-sourcing), there are still a plethora of application types to fill in the plant-level detail. Age-old Material Requirements Planning (MRP) and Capacity Requirements Planning (CRP) determine unconstrained combined raw material requirements and a rough estimation on plant capacity utilization respectively. Usually the input for either is an even rougher Master Production Schedule (MPS). CRP evolved into Finite Capacity Scheduling (FCS) which has the same objective, but does it accurately, and combined with MRP evolved further into Advanced Planning and Scheduling (APS) to cover both material and capacity simultaneously and accurately. A number of industries also have a need for formula optimization, where the mixture of individual lots of raw materials or intermediates are determined to produce other intermediates or finished goods within certain specifications. For beverages these specifications may be sugar content and acidity; for chemicals these may be elemental specs or particle size distributions. This falls squarely within the SCM domain. In various scenarios the end specification may be met by choosing from a named number of predefined recipes, rather than the infinite options of mixing any amount from any lot with one another. In those scenarios a simplified version, commonly known as recipe optimization may be used. Some SCM and some ERP systems provide the latter, although strictly speaking this too should be owned by SCM.

Supply

Finally the pull of demand reaches the start of the supply chain via all the areas mentioned above (not necessarily in that order). For most businesses there is little at an operational planning level on this side of the supply chain. At a tactical level, most businesses incorporate a sourcing decision for particular suppliers. But once you know the requirements from manufacturing it boils down to getting the desired product at the desired quality at the lowest price, which is left to the purchasing department and formalized using ERP. Some industries however have to deal with seasonal supply; for example when the raw material is harvested. This seasonality has an impact not only on quantity available, but also on quality of the product and price. For these businesses it is important to forecast supply in similar fashion to forecasting demand. Both these forecasts would input into a Supply Chain Planning (SCP) optimization to determine optimal tactical plan to minimize overall costs.

Holistic

Besides the point solutions mentioned so far, there are also a number of holistic planning or optimization areas within SCM. SCP mentioned in the previous paragraph is one such. Along with demand forecast and possibly supply forecast it incorporates aggregate data on manufacturing, warehousing and transportation capacities, to come to its results. The results of these strategic and aggregate optimizations are the inputs to the various point solutions. The results could include suggested shift patterns or overtime, changing routes from default sources, along with the standard capacity allocations. These are all tactical planning decisions, although in many businesses the term “strategic” is used for some of its results (such as stratetic inventory buildup). Another area that potentially spans the entire supply chain is Capable to Promise (CTP). Where ATP looks only at demand forecast consumption (or only at on hand inventory consumption), CTP looks at available manufacturing, warehousing and transporation capacity to give a promise in certain cases where ATP cannot. An extension of that is Profitable to Promise (PTP), where the incremental costs of the freed up capacities are taken into consideration to determine if it is worthwile to promise even if there is a capability. A popular topic within SCM that spans multiple areas is Sales & Operations Planning (S&OP). This is very treacherous ground where representatives from sales teams and various operations departments battle over what is possible. Historically this business process requires all the team members to meet in a single room on regular basis to come to agreement on how to compromise mismatches between what sales has promised or wants to promise with what the supply chain can cope with. There is great potential in relieving this conflict using SCM software. Finally, there are some strategic optimizations of which Strategic Network Optimization (SNO) is the most formalized one. This optimization determines whether it makes sense to open or close facilities or reduce or increase capital capacity such packaging or production capacity in individual facilities. Outputs of strategic optimizations can be used when entering into long-term contract negotiations, rather than be constrained by them. Most other strategic optimizations are ad hoc custom solutions built for a very specific purpose.

Well, if you are reading this line, you got further that I thought anyone could come in this rather boring summary of what SCM entails. This was a difficult piece to write because there is just so much to say about each area. I sought to find a balance between clarity through extent of descriptions and providing examples and brevity to the point of significance to the real purpose of this blog: the future of SCM. The end result is more tedious than I like to associate myself with. But so be it. If popular demand persuades me to write in more detail about certain areas I may do so, but until then I will happily lay the past and present to rest and focus on the future.