Infosys’ blog on industry solutions, trends, business process transformation and global implementation in Oracle.

« Infosys accelerates application integration and speeds business value realization through Oracle PaaS implementations | Main | HR-bots and baby day-care centers: How today's Enterprises are balancing their HCM initiatives »

Customer Delivery in Semiconductor Industry

Introduction

Semiconductor products finds application in various industry verticals that requires digital technology embedded in their end products to enrich them with the ability to program, configure, connect to other devices, automate or provide extra features to delight the customers.  Analog semiconductor application are also diverse such as in the field of power electronics, servo control application and energy.  Some examples are machine control units in automobiles, smart mobile phones, and control units in industrial automations, robotics, solar energy etc. 
With a wide range of application, semiconductor industry commands a business little more than $330 billion worldwide.  Supply chain in the semiconductor industry is characterized by the price sensitivity, constant innovation, product complexity, collaboration with manufacturing partners and globalization of its value chain.  The semiconductor product is not an end product by itself and is usually used to build one and hence its supply has a significant impact to the business downstream.  Low margin in this business requires higher volume turnover for them to be in green.   Competition and the above challenges forces the industry to emphasize on the customer delivery performance and service levels to their clients, who are from diverse segments in the market. They have to add value not only to the technology through innovation but also to the supply chain process by reduced cost along with reduced reaction time to demand.


  Customer delivery function is identified as the priority one processes to sustain in the semiconductor business.  It involves order capturing, calculating demand, delivery promising, tracking logistics till ownership transfer to customer and keeping customer informed.  This is supported by the planning function to ensure service levels that reduces or prevent line down situations for their clients and maintain optimum inventory.  The key performance elements of the customer delivery that depends on the supply chain are -
a. Reaction time to the demand.  The ability to respond with promises or exceptions to the client.
b. Ability to meet the customer demand as per the schedule with negligible or no slippages
c. Real time status information availability.  This is the ease of accessing status of inventory in terms of location, reservation and quantity.





Processes for improved Delivery and service levels

Customer delivery is handled by conventional method such as receiving request for quotation, sending a quotation and then confirming delivery using sales orders. However this method is not suitable for high volume staggered by demand schedules over a period of time. This necessitates the reduction of negotiation time and collaboration with the supplier to meet the production schedule.


The continuous demand at different schedules is handled by the schedule agreement processes, which is a method where the customer and supplier have an agreement for supply of a product, at a particular price and to a ship to location.  The quantity and time of delivery is decided by the forecast and firm demand that is fed by the customer to the supplier. Forecast demand provides the future demand visibility and the firm demand is used to execute the delivery.  


Supplier plans the supply based on the forecast demand and promises the customer with the delivery dates so that they can plan the production as per the promised availability.  This demands daily planning executions so that exceptions are resolved in advance and promise/commit dates are provided at the least possible time.  The promise dates once committed is maintained or bettered when the subsequent demand schedules are fed to the supplier's system.


Once promised the supplier should be in the position to have the item available for the customer's firmed demand. The firmed demand is usually within a demand time fence and is not allowed to change because any further adjustment may not be possible due to the item's availability lead time. Theoretically the quantity reduction will allow the quantity to be freed from a schedule and reserved for the next schedule.  Increase in quantity can be met only if there are freed quantity from other schedules of the same customer or another customer.  


There could be scenarios where the customer's production schedule cannot follow forecast demand strictly and need deviation from the forecast within allowable tolerances.  Further the availability of components should meet the firmed demand which does not match the forecast. This scenario is seen in auto industries where the forecast may not match the orders.  The production schedule is dependent of the orders on hand due to the high cost products.  Consigned inventory process is used in such scenarios where the safety level is always maintained at the customer's premises so that uncertainty and item availability does not compromise the production schedule of the firm demand.  The replenishment to the consigned inventory is calculated by the supplier based on the forecast by the customer in the vendor managed inventory processes.  If the customer plans the replenishment then the customer managed inventory process where the supplier ensures the supply meets the schedule suggested by the customer.   In both the cases the supplier owns the inventory till the customer consumes it in their production.


Unlike automobile industry, the customers who manufacture mobile phone and other low cost products do not plan the production purely on the orders on hand.  The historical sales of the same product or the similar products are used to determine the production schedules.  This scenario will not need the consigned inventory process. However the supplier will need to meet the demand as per the schedule transmitted by their customers.  Just in time direct delivery processes as per the pre- negotiated agreement is used.  Forecasted demand is used by the supplier to plan the supply in advance and replenish their inventory by not only ensuring reduced inventory cost, but also ensuring the item availability for delivery as per promised date.  Once the demand is firmed the delivery is done.


The above two processes ensures low or no slippages on item availability and hence improving the customer delivery function.  The third key element is the real time data availability for the customer delivery personnel or to the customer themselves.   This requires the data in the supplier's system to be real at any moment of time.  Following processes impacts the real time information.


a.  Replenished quantity in the warehouses is impacted by the real time receipt transaction execution. This is  required for offline planning by the customer delivery personnel
b. Replenished quantity in the consigned inventory is impacted by the delivery receipts at the customer premises and the consumptions by the customer.  These net inventory in the consigned inventory is reconciled regularly as this is used for the replenishment calculations and in transit quantity calculations.  Having accurate date at consigned inventory is a challenge and effort is made to collaborate with the client to reconcile them regularly and to form a mechanism to collect receipt and consumption data with the least lag.
c. Semiconductor products are lot controlled.  Hence having the information of lots at the consigned inventory that are not consumed is a challenge unless there is an agreement in the method of consumption.  First in First out (FIFO) or Consume the lowest expiry date lot process is used to have the lot information as accurate as possible.
d. CUM management is a method to keep track of cumulative deliveries for an agreement. This is reconciled with the data provided by the customer to arrive at the in transit and quantity at the customer location.

Above are the revenue generating processes.  Other implicit processes to improve the service levels addresses the handling of defects, its replacement etc.  Some of these have process have financial implications where the invoice generated is reversed using credit memo.  The return of the products that are sold are returned using the return material authorization processes (RMA). That results in the credit invoice after receiving the product back.  Consigned inventory products are not sold even though they are at the customer location. Hence the defective consigned products that are not sold are transferred back to the source warehouses using the inter organization transfers without impacting the financials obligation of the customer.   Consigned process such as CMI and VMI also need to address any wrong consumptions.  In that case the consumption reversal that transfers to the ownership back to the supplier is used. This process is similar to the RMA but the destination would be the consigned inventory rather than the source inventory.


 Operations for improved Delivery and Service levels

Ease of functioning is important for a particular process to be successful.   Real time visibility of status for every transaction,   ease of bulk uploading of the voluminous data, automated demand planning, automated supply planning and elimination of irrelevant data considered as noise are important considerations while implementing a processes for improved delivery and service levels.   The risk of functional failures is mitigated with alternatives such as manual process to prevent business stoppers.


 

Ease of data input is achieved by the electronic data interchange (EDI) and bulk upload methods.  This eliminates the need of the user to input the data manually.  With the customer who do not have EDI connectivity, the data is gathered in the spread sheet and the same uploaded to the system.  To avoid risk in event of the functionality failing the manual entry is provided.  These methods need the demand data to be directed to the required to agreement specific to a product, ship from and ship to locations.


Automation of the data validation, then processing it to generate the calculated demand based on the process used and present the planned orders for user approval before progressing for execution are the features that will eliminate the user's effort for mundane activities and eliminate human errors.  Exception handling, transparency of demand calculation or supply calculation and ability for the users to review it before executing the plan are necessary to build user's confidence on the system.  


 Visibility of the relevant information such as stock availability globally, real time status of deliveries and real time exceptions enables quick responses by the users to the situations like taking quick decisions for line down, keeping customers informed and resolve data issues.   Reporting of system errors and exceptions due to wrong data cannot be delayed as that would provide the false impression that all is well and resulting in wrong expectation. Hence provision for real time reporting of such is important. However only the relevant details required for the owned actions by the user is reported, hence reducing the effort of identifying.    CUM management is a method to track the cumulative deliveries calculated from a start date and hence is used to calculate the in transit quantities, review historical deliveries and for reconciliation etc.  CUM management provides visibility of the current and past transactions.  Customers providing the CUM quantity can be cross verified and deliveries reconciled.


Functional Landscape- Using Oracle Applications

Semiconductor business is viable for large business due to its high capital cost for manufacture and logistics.  Hence the enterprise solution revolves around the assumption that transaction are of high volume, automated planning and lesser fulfillment time.  

  
 

Above is the illustration of the various functional modules that are necessary for a fairly automated customer delivery process. Oracle Finance and Manufacturing modules are separate modules for implementing finance and manufacturing processes. Nevertheless the support from these are necessary for successful customer delivery. The key features of these modules are


a. EDI gateway-  This enables interaction with the trading partners ( suppliers and customers) using EDI and hence automating the transfer of data related to delivery orders, advance shipment notice, commit dates, change orders, forecasts, invoices etc.
b. Release Management- This module directs the incoming data feed such as forecasts and firm demand to planning and order management based on processing rules.  Processing rules are set for specific to ship to, ship from, ship to address and item.
c. Demantra is used for demand planning based on historical sales data, incoming forecasts/ demand and S&OP (sales and operation) planning.  This demand is fed into the advance planning modules
d. Collaborative planning enables planning by collaborating with the suppliers and customers. The demand visibility is shared to the supplier who in turn plans his production schedules. Vendor managed inventory and consigned inventory process are implemented by this module
e. Order management is necessary to capture firm orders that are to be delivered.  This has enough features to track the deliveries and allows the users to override delivery schedules.
f. Purchasing- This is used for generating requisitions for transfers within the organizations.  Planning engine will generate recommendation for purchases from external vendors. When released it will create purchase requisitions and purchase orders for the procuring in this module. Other processes like global procurement, drop ship, outsourcing, consigned for vendor are configured using this.
g. Advance planning is used for supply planning based on the bill of materials, supply network, routings, distribution network and demand.   Planning is tweaked based on required key performance indicators and planners can override wherever required.
h. Warehouse and inventory modules are used to implement the picking, shipping processes. It provides the global visibility of the inventory.
Key processes and operational consideration in the solution includes
a. Automation:   Automation or ease of demand data input to the supplier system and automated demand and supply planning.  Planners and customer sales representatives of the customer delivery personnel have provision to override and handle exceptions.
b. Transaction standardization:   Have agreeable standards for transactions that can be automated.  For example consumptions and delivery receipts at the customer location if not communicated within a specified time is transacted automatically.  Another example is release for shipment if the commit date or the promise date is within the demand time fence.
c. Mitigation:   Process for mitigating functional failures such as planning engine unable to run. Then the planners should have provision to provide the promise dates.   Similarly failure to automatically release the order for shipment will have provision for manual intervention.
d. Reporting:   Have on demand report generation solution for inventory status, demand and supply information, transaction information and ownership of account information.


Conclusion

     Semiconductor market being around $330 billion is not expected to show significant growth in the near future.  The individual players will compete with better services to their clients and by enhancing customer delivery.  Automation and ease of operations along with better service levels is necessary for cost reduction and customer loyalty.  Standard packaged applications such as Oracle E business suite and Oracle fusion cloud application provides industry's best practices.  Individual organization adapting to these processes focusing on customer delivery will have the quick turnaround to become competitive in the industry.  This will drive the streamlining of downstream processes.


Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on

Blogger Profiles