Supply Chain Management: Supply Chain Traceability in the Automotive Sector: Need for “back-to-basic” solutions

As automotive sector carries on through the roller-coaster ride of plummeting and rising sales, the regulations which impact the industry are growing stiffer by the day. These regulations insert a lot of pressure on manufacturing and supply chain functions to maintain clear track of the operations, value-adds at each stage and the ensuing responsibilities of each stakeholder in the automotive value chain. As cost pressures on today’s scenario are foreboding on these regulatory pressures, OEMs (Original Equipment Manufacturers) and suppliers would be least interested to spend any extra dollars on establishing any new traceability solutions. This leaves most OEMs and Suppliers vulnerable to traceability gaps in their supply chain which enhances their risk exposure. In such circumstances it is imperative that OEMs and suppliers return back to the basics to device low-cost traceability ideas from their legacy infrastructure.

Supply chain traceability, for ensuring flawless tracking of every small part or process in the supply chain, is required basically for three main objectives:

Ensuring quality tracking and feedback in the supply chain
Discover hidden inventory in the supply chain to optimize its size
Discover process bottlenecks which can be eased or eliminated to improve operational efficiency

Here we deal with the issues faced in the supply chain and their impact on the objectives mentioned above. It is to be noted that Operational Line Managers do not subscribe to hi-tech methods for gaining clear visibility into each aspect of tracking in the supply chain. Most Managers believe in “gross tracking” of products and processes. What this means in a real world scenario is that the accounting of the volumes during product or process hand-over is on a broad gross level and batches are identified as single block. This procedure is mainly adopted either in Tier-2 or Tier-3 supplier level or at child component production level at the Tier-1 suppliers. Instead of identifying individual child components, the complete batch is identified. E.g. batch of heat treatment, batch of molding, batch of casting pouring, batch of machining. There are three critical aspects to be weighed sensitively for adopting a particular process for batch identification.

Identification Technique: For such batch identification the technique adopted for identifying the batch and its error-proofness (also referred to as pokayoke by Japanese manufacturers) is justified by the calculations of risk potential for the identification failure. E.g. Heat code present on Cam shaft casting provides identification of the Pallet# of cast, Ladle# of melt, date and shift of pouring etc.

Identification registry: For such batches with identifications, the recording or registry of these identification marks either permanently on products or in some discrete records is essential for any future retrieval. In case of any future mishaps, this record/ registry enable accurate “search and locate”, to ensure a thorough quarantine operation. The justification for the cost of recording, retrieval and archiving for a predefined period is purely based on the risk potential of the record failure. E.g. Permanent marker identification mark on the back of MFR (multi-focal reflector) of headlamp after Al-evaporation process is recorded and maintained during an 11-year period for any future photometry recalls of vehicles.

Risk Potential ($) = [Frequency of occurrence (n)]

X [Cost of Impact ($)]

X [Risk resistance multiplier (λ)]

X [Scope of failure detection (δ)]

In this equation the Cost of Impact is calculated in $ terms and is directly derived from the cost of mitigation of risk and the cost of lost revenue due to reputation damage. On similar lines frequency of occurrence is calculated from historical data or from Conditional probability tables (CPT), and also depends on lot sizes and the number of lots produced. Risk resistance multiplier is basically derived from the confidence index of the occurrence probability and the type of distribution that the failure trend has best fit with. Scope of failure detection is another multiplier which denotes the ease of detection of the failure when the control mechanism has failed.

Integration of batch identifiers: This is the section where the complete supply chain comes into picture. Whereas the above aspects of traceability are a simple cost-benefit analysis, this integration poses serious operational challenges to the Line Managers. Since batch sizes of sequential operations are different, and many a times disjointed, maintaining this traceability in the supply chain poses a serious threat. E.g. 37 lot size and 56 lot size castings of Brake-drum is followed by a 86 lot size and 102 lot size of surface finish machining operations. This causes the second lot of castings to be split into two machining operation lots. To add to this complexity, if the yield from these lots is less than 100% due to rejects/ damage/ trial scrap then the situation becomes very cumbersome to handle. To further complicate the matters, any supply chain logistics among multiple plants where the castings and machining are to be warehoused and transported, then the storage bin quantity, inter-plant order quantity, transportation lot size add other dimensions of complexity. To what extent the critical identifiers are recorded and their records maintain is a “grey area” that most Supply Chain Managers dread to tread.

It is these types of challenges which pose the most serious threat to Manufacturing and Supply chain managers. So most supply chain managers would feel comfortable with single-piece flow, but operational and cost constraints force batch processing. Some risk-averse managers would resort to recording of details of each and every lot that passes their gate, and if costs are overbearing they would impress upon the Top Management to make “necessary” expenditure in RFID and the like. But most Line managers take a pragmatic approach. They either design batch sizes which are simple multiples or whole-number fractions of preceding lots or the lots to follow. This enables them to keep track of only a few records kept at hand-over points. In may manufacturing and supply chain sites one can observe lots of 24 or multiples. This is so because 24 have the most factors of single digit even numbers and this eases a lot-split decision. Do you also face similar supply chain traceability issues and what cost-effective solutions do you deploy?

Comments

One of the main problems with the Automotive Supply Chain is that of inventory. $m's if not $bn's of unsold cars litter the World. Ultimately these car graveyards keep prices high and margins slim. Moving towards build to order where supply chain's respond to consumer demand in minutes (not months!) is the key. It is possible to achieve this, through improved forecasting, end-to-end visibility and real-time replenishment (to name but a few initiatives). Roll-on Ford, GM & Toyota seeing the light. Kind regards Matthew @ Perceptant.com

Posted by: Supply Chain Management Doctor | December 15, 2009 5:22 PM