Supply Chain Management Systems for Service and Replacement Parts: Players, Benefits, and User Recommendations

The differences between new parts production supply chain and service and replacement parts supply chain are significant. Companies using conventional supply chain management (SCM) methods to track their service and replacement parts supply are failing to grasp the special needs of the aftermarket. Further , one can even differentiate between the inventory optimization approaches of new production parts. Pure distribution parts include finished consumer goods (not including fashion/apparel items due to their seasonality idiosyncrasies, see Intentia: Stepping Out With Fashion and Style; Part One: Characteristics and Trends of the Fashion Industry), with a large number of items, large number of locations (whereby store levels can get out of hand), and with a desire for very high customer service levels (98 percent or more). The vendors that cater to these customers would be the likes of ToolsGroup. Mixed manufacturing and distribution for new parts require the exact positioning of parts. Exact positioning is highly important for manufacturing and configuring (postponement) purposes, because bill of materials (BOM) logic is heavily leveraged for inventory planning and optimization. Leading vendors in this market include Optiant, LogicTools, SmartOps, or i2 Technologies.

These solutions typically leverage stochastic optimization using nonlinear modeling techniques to analyze input data for randomness. This knowledge is applied to determine an optimal inventory policy at a particular node in a multi-tier supply network. Namely, as supply chain variability has increased, the data has become more random. Consequently, user companies need to not only look at the nominal values per se, but also at the probability of the value. For example, they need to know what the probability is of the forecast value, the purchase/transportation lead time, the manufacturing run time, the supplier quality, etc.

Contrary to these new parts production segments, aftermarket service and replacement parts are typically "slow movers," but may be critical for the operation of expensive equipment, often containing associated service level agreement (SLA) penalties for inadequate service. Thus, for the reasons of repair and indenture level and SLA considerations, one has to optimize inventories for required service levels and end-equipment availability. Varied service and customer entitlements complicate things, since aftermarket service must support warranty commitments; contract extensions, which might include same-day or next-day service; and direct or through distributor part sales. These entitlements may have different service objectives, which may include fill rates, response times, or system uptime maintenance.

How is risk factored into decision-making for service parts? In this case, forecasting might use demand history, but perhaps more importantly, mean time between failure (MTBF) data and an analysis of causal factors can provide item- and location-specific estimates of usage. This data can also be used to calculate the probability of demand occurring during the planning period in question. For example, a forecast might state that there is a 12 percent chance that the user will need a specific part in the next thirty days at a specific location. However, this forecast is risk-based, rather than consumption-based, as it is in new parts production supply chain planning (SCP).

Moreover, the design of the distribution network including which parts and how many of each are positioned at which depot(s) is another risk based evaluation. Typically, this dictates a multi-tier or multi-echelon depot strategy, where tactical planning involves risk-based decision-making that considers the probability of demand and therefore, the probability of a stockout. To refresh our memory, stockout costs may include lost sales, backorder costs, expediting, and additional manufacturing and purchasing costs (not to mention lost face before the customer and hurting SLA penalties). Thus, the strategy include issues like, if we have a 20 percent chance of needing a single unit of a specific part in the next thirty days, what are the odds that we will need two? Moreover, given the part delivery lead time, what are the odds that the demand for two will create a stockout?

SOURCE:http://www.technologyevaluation.com/research/articles/supply-chain-management-systems-for-service-and-replacement-parts-players-benefits-and-user-recommendations-18090/