4. Case Study
4.1 Case1 : e-supply chain BPR
Founded in 1987, TSMC was the world’s first pure foundry, focused solely on the manufacturing of semiconductors. Operating in the cyclical semiconductor market, the company managed to grow rapidly and to become the world’s 5th largest semiconductor manufacturer with more than 50 percent market share in the foundry business. Before emerging of TSMC, all the Integrated Circuit (IC) manufacturers in the semiconductor industry were Integrated Device Manufacturers (IDMs)-fully integrated manufacturers that designed, fabricated, and marketed their own products. However, by the mid-1970s many had started outsourcing portion of their manufacturing activities. In the mid-1980s, entrepreneurs with new ideas for IC chip designs searched for IC fab to manufacture their
products. These chips companies became known as fables organizations that concentrate their engineering resources on innovating new product designs rather than on product manufacturing. The market demand from IDM’s outsourcing and fables companies resulted in the disintegration of the semiconductor industry and creation of foundry business model.
From 1998 to 2004, TSMC and ASE conducted a major BPR project by completing electronic process integration of 11 key business processes through the Internet. They can now obtain accurate, timely information on their product status. The direct economic benefits are estimated to be around US$ 10M through productivity increase over a total investment of about US$ 2M; the indirect benefits of this initiative could be on the order of US $100M million if the joint customers’ benefits are considered. In collaboration with the RosettaNet organization, TSMC and ASE leveraged their BPR experiences to define three data exchange standards that can then be widely adopted in the semiconductor industry.
4.1.1 Initiation stage
(Challenges and Problems) The continuing trend of the semiconductor value chain disintegration has resulted in specialized companies that are independent yet co-dependent upon one another, as shown in Figure 5. These companies need to closely collaborate with their partners both up and down the value chain to meet the relentless end-consumer demand to achieve shorter time-to-market, lower costs, higher responsiveness, and better quality. Therefore, streamlining the business processes between partners, sharing information appropriately, and ultimately effectively “re-integrating” the value chain in a virtual manner have become the most critical issues in the semiconductor industry.
Figure 5 Disintegration of the Semiconductor Industry Value Chain
(Review company’s vision) The vision statement of TSMC is “to be the most advanced and largest foundry services provider to our target customers, and in partnership with them, forge a powerful competitive force in the semiconductor industry. To realize the vision, we must be:
(1) a technology leader, (2) the manufacturing leader
(3) the most reputable and service-oriented and the greatest total benefits provider.”
In the company’s early days, TSMC focused on manufacturing excellence and technology leadership. As competition in the sector intensified in the late 1990s, the company began to focus on customer service to further differentiate itself from competitors. Therefore, in order to respond to the industry trend of value chain integration and strengthen its market competition, TSMC decided to implement the concept so-called “Virtual Fab” (customer can online engage business and inquire about the status of their production orders as if they were in their very own manufacturing unit) to transform itself and become a service-oriented enterprise. The concept of “Virtual Fab” would be realized by a series of initiatives defined in a strategic “e-foundry” program. As shown in Figure 6, the
“e-foundry” consisted of three initiatives: design, engineering, and logistics collaborations, which as a whole integrate TSMC and its upstream customers and downstream partners in testing and packaging. All the major business activities between TSMC and its customers, and its downstream partners in testing and packaging that contribute the IC manufacturing and delivery life cycle are all included. In order to meet different needs of information access by different kinds of customers, TSMC decided to establish two types of information delivery channels. One is TSMC-Online that provides web access interface to allow all customers manually access the Design, Engineering, and Logistic related information; and the other is TSMC-Direct that provides even faster and more reliable system-to-system link to selective customers with packaged information.
Figure 6 Illustration of TSMC e-foundry
(Identify the key processes and select the ones to be re-engineered) While preparing to implement the “e-foundry”, TSMC identified 15 main business initiatives and/or processes in the categories of Design, Engineering, Logistic Collaboration and Common Foundation as shown in Table 8. To define the implementation priority, these 15 business processes and/or initiatives were compared according to the level of implementation effort and their business impact. As shown in Figure 7, the process of “Tracking of service order status”,
“Delivery of engineering specification”, and initiative of “Definition of standard data
Assembly & Test Partners
exchange” were identified as quick win owning to their relative low level of implementation effort and high business impact. In order to provide the complete set of information including both IC manufacturing front end (wafer manufacturing) and back end (wafer testing and assembly) together to customers, TSMC decided to integrate the supply chain process with its major partner, Advanced Semiconductor Engineering Inc. (ASE), the world’s leader in semiconductor assembly and testing. These three process as a whole would serve the foundation of TSMC-Direct.
Table 8 eSCM preliminary business process initiatives
Category Key Process/Initiatives
Common Function Componentization of TOM / OTIS / Yes views Definition & implementation of standard data ex Definition and implementation of data adapters
Provide enhanced granularity of WIP reports
Emerging technology exploration
Design Collaboration Customer feedback engine
Self directed needs assessment and process selection
Decision support solution configuration
IP Repository
Logistic Collaboration Integration of 3rd party web sites (e.g. design enablers, test / assembly houses)
Tracking of service order status
Proactive notification of updates, bugs, and fixes Engineering Collaboration Self directed troubleshooting guides
Live “classroom” virtual training
Electronic delivery of Engineering Specification
Figure 7 eSCM process Impact and Complexity Matrix
(Set BPR objective and designate process owner) The objective of the TSMC/ASE e-Supply Chain integration process was to integrate key operational activities and data between TSMC and ASE, resulting in a seamless information and transaction interface to their joint customers, as if manufacturing took place in the customers’ own backyard. Specifically, the objectives of the strategic business process were:
Streamline and integrate key business processes
Improve transparency of information exchange
Increase speed of information delivery
Standardize process “hand-shaking” protocols and data exchange formats.
Due to the process nature is to integrate supply chain via the modern information technology, a senior VP of Cooperate Development Organization in charge of Information Technology and process integration was designed as the “efoundry” program executive officer; and process owners were jointed performed by the senior managers responsible of e-Commerce in TSMC and ASE respectively.
(Search for BPR benchmark target) As the pioneer to integrate the supply chain process in
High
Low High
Business need/Impact
Level of Effort
• Integration of third party web sites / extranets (e.g. design enablers, test / assembly houses)
•IP Repository
• Electronic delivery of Engineering Specification Forms
•Provide enhanced granularity
of WIP reports •Self directed needs assessment and process selection and
service order status •Proactive notification of updates, bugs, and fixes
•Emerging Technology Exploration
Source: TSMC
the semiconductor industry, TSMC and ASE could not find reference in the same industry but instead to search for the successful case in other industries. Amazon and Dell were identified as the benchmark targets due to their excellence in the following aspects.
Planning/Inbound Logistics
eProcurement
Configuration
eJIT/Inventory Management
New Product Introduction
Outbound Fulfillment
These two companies not only demonstrated robust and agile new product introduction and demand fulfillment capability, but more importantly, the e-commerce capabilities they possessed help them to create competitive advantage and secure the leadership in the industry they operated.
(Secure commitment from top management) As shown in Figure 8, TSMC’s competitive advantage had been evolved from manufacturing-focus in 1987,technology-focus in 1997, to service-focus in 2001. As part of service-focus total solution, this TSMC/ASE e-Supply Chain integration process had the strong commitment from top management undoubtedly.
Figure 8 TSMC strategic focus evolutions
4.1.2 Study
(Organize the BPR team) As a cross-companies project, the BPR team was jointly formed by TSMC and ASE respectively with team members from business and IT organizations involved in the process. For TSMC site, the key organizations involved in the business process with ASE were “Backend Technology & Service” and “Manufacturing Product Engineering”; for ASE were “Production Planning” and “Manufacturing Product Engineering”. These organizations played the role of process review and redesign from the business operation perspectives; whereas, the IT organizations of E-Commerce at TSMC and Service Integration at ASE respectively played the role of process automation and system implementation. Since this e-Supply Chain integration project was fresh for both companies, an outside consulting team was hired to share its successful experience of other companies implementing the similar processes. The figure 9 shows the organization chart of the BPR team.
Figure 9 e-SCM BPR team organization
(Delineate, measure, diagnose the existing process and propose the concept the of new business process) There were plenty of manual activities between TSMC and ASE business process before the BPR was conducted. Take “order process” as an example, as
TSMC Team AST
shown in figure 10, there were totally 10 steps for TSMC and ASE to complete a work order cycle. Not only long processing cycle time, the manual process also caused unexpected operation errors and in turn impacted the accuracy of billing and manufacturing planning.
After thorough analysis, the BPR team identified 4 key elements to complete an order process; i.e., PO release, Available to Promise (ATP), PO conformation, and PO acknowledge. The BPR team then proposed the concept of replacing the repetitive manual activities with an automate solution by a business-to-business linkage. The manual process would only be activated when there were exceptional cases that were beyond the definition of 90% of ordinary order type. The Table 9 compares the before and after the redesign based on the seven rules described in the previous chapter. The BPR team also identified three indices to measure the business processes; i.e., processing time, data transmission time, and error rate.
Figure 10 As-is and To-be analysis of eSCM Order Process
Table 9 Before-and-after comparison of the e-SCM BPR
(Ask feedback from user groups) After introducing the concept of the new business process to the user groups of the two companies, the feedback was very positive because that the user groups were suffered for a long time the manual process of numerous amounts of business transaction and data transmission. According to the user group of logistics, there were around 20,000 work orders every year, and the average error discover and recovery time was 40 man-hours. Since the full logistic cycle included “Delivery Schedule Planning”, “Work Order”, “Order Confirmation”, and “Shipping/Invoicing”, the user group requested the B-to-B solution should incorporate all those processes as a whole.
In addition to the logistic data, there were equivalent amount of data need to be exchanged to serve the engineering purpose. The data required by he engineering user groups included engineering specification, testing results, yield data, etc. In order to maximize the benefits out of the B-2-B solution, the user group suggested the e Supply Chain integration project should cover the scopes of both logistic and engineering.
(Design the prototype and ask feedback from user group) As the concept and scope of the BPR were decided, a prototype of the “Work Order (WO)” was developed to demonstrate proposed redesigns and to collect rapid feedback that would help estimating and planning
Before After
Rethink (why) Reassign (who)
manual process by business operation staff
primary by system secondary by manual Reconfigure (what)
all activities for normal and exceptional cases
system handle normal cases manual handle exceptional cases
Resequence (when) evet trigger defined time and frequency
Relocate (where) no change no change
To achieve shorter time-to-market, lower costs, higher responsiveness, and better service quality, it's critical to integrate the value chain of semiconductor industry
detailed business and systems requirements. After reviewing the high-level business processes, execution flow and algorithm of the prototype, the representative user group found several design flaws. For example, a “Work Order (WO)” process was defined as a sequence of activities: TSMC sends a WO to vendor, ASE receives the WO and sends back an acknowledgement to TSMC to complete the purchase order. However, when a transaction failed, the prototype system could not tell whether the failure was caused by inconsistent data content, application server malfunction, or a network problem. To cope with this disconnection, a two-way acknowledgement protocol and error-handling procedure should be implemented. Such valuable suggestion was quickly accepted by the BPR team and put into formal implementation plan.
(Develop change management plan) Since the ultimate goal of the supply chain integration project is to achieve shorter time-to-market, lower operation costs, and better service quality, both the management and operational level of TSMC and ASE held high expectation of the project success. The focus of the change management of this project was not resistance mitigation of the user groups impacted by the new process but the smooth transition from the manual to automatic process. The BPR team analyzed the key concerns of the project stakeholders and defined corresponding change management plan.
As shown in Table 10, management and operation had common and distinct concerns from their perspective respectively. The management team was more concerned about the project schedule, cost, quality, and results; whereas the operation team was more about the operation effectiveness, such as system reliability, exception handling, and continuous support. Nevertheless, all the concerns were addressed properly in the change management plan.
Table 10 e-SCM change management plan
Concerns Change management activity
Management Project delivery time
Project cost
Solution quality
Customer satisfaction
Define clear project milestone
Monitor project progress
Periodical project status update
Measure performance to ensure customer satisfaction
Operation Project delivery time
System reliability
Exceptional case handling
Continuous IT operational and enhancement support
Define clear project milestone
Collect end user requirement
Define standard operation procedure
Build strong IT infrastructure
(Report BPR study result to senior management) After reviewing the BPR study result, the senior management team approved the project scope, budget, and implementation plan. The project was comprehensive in scope, encompassing all major business activities between TSMC and ASE in the following two dimensions:
(1) Engineering Collaboration, including – Engineering specifications – Boding diagram
– Yield data
(2) Logistics Collaboration, including:
– Work Order and Order Response – WIP data and WIP Tracking – Event and Inventory
– Finished Goods and Ship Out Data
It’s estimated a two-year project with US$ 2 million dollar budgets for purchasing hardware, software, and external consultant. Additionally, it would also require 12 full-time-equivalent TSMC and ASE employees to conduct business requirement analysis, system design, and implementation.
4.1.3 Pilot
(Develop Pilot process and small scale implementation) After numerous mutual visits, meetings, e-mail exchanges, and telephone communications, the business-to-business integration of 11 e-processes were established including yield rates, testing results, order and order acknowledgement, work-in-process, and shipment of finished products in stock;
etc. Figure 11 illustrates the identified key processes between these two companies. In order to achieve quick win and prove the concept of B2B integration, five of the eleven processes were chosen to be implemented in the first release due to their relatively low complexity or data volume as shown in Table 11. Those processes were Assembly/Final Test Engineering Spec, Bonding Diagram in Engineering category and Work Order, Work Order Response, and WIP in Logistics.
Figure 11 TSMC/ASE’s key process integration–conceptual overview.
* AS: Assembly, FT: Final-Test, ERP: Enterprise Resource Planning, MES: Manufacturing Execution System Pilot scope
(Source: TSMC)
Table 11 Complexity and volume comparison of 11 eSCM processes
(Ask feedback from user groups and modify pilot) The result after implementing the pilot process was very impressive. As shown in Table 12, the data transmission time and transaction processing time were improved 5 to 6 times and the error rate was significantly decreased 8 times.
Table 12 Before-and-after comparison of eSCM Order process
Given such encouraging result, there were still some process misalignments being identified by user groups. For example, one-to-many relationship of customer’s product name to vendor’s internal manufacturing device name often caused confusion when both sites handled order taking and invoicing. Consequently, a significant amount of manual effort was placed to correct operation errors. To fundamentally solve this problem, the project team re-defined the product’s naming scheme and generated a mapping table to maintain a one-to-one relationship.
(Report BPR pilot result to senior management) The success of the pilot proved the concept that the value chain integration could result in closer collaboration between companies.
After reviewing the satisfying pilot result, the senior management of TSMC and ASE
Items Before After
Data transmission time 120 minutes/order 20 minutes/order Order Processing time 100 minutes/order 20 minutes/order Order Errors 12 orders/month 1.5 order/month Dimension Process Complexity Volumn
Eng spec L L
Bonding diagram L L
Yield data H M
Work order L M
Work order response L M
WIP L M
Event data M L
WIP tracking data H M
Move into inventory data M M
Fisish good bank data H H
Ship out data M H
Engineering
Logistic
determinedly select next wave of collaboration partner both upstream and downstream of the value chain as shown in Figure 12. The top management from both TSMC and ASE also set an ambitious goal as setting the industry standard with the data exchange formats defined by the two companies. It was meant to be the “foundation” upon which more process integration and data exchange would be established both upstream and downstream the entire semiconductor industry value chain.
Figure 12 e-SCM: extending integration up and down the value chain
4.1.4 Full scale implementation
(Define the full scale BPR development plan): The full scale development plan consisted of 2 phases and three distinctive milestones as shown in Figure 13. The phase 1 would build 11 e-processes integration between TSMC and ASE in two years. The phase 2 comprised of two tasks in parallel, one is to extend the B-2-B supply chain integration solution to both upstream and downstream partners of these two companies; and in the mean time, submitted the data exchange format defined by TSMC and ASE to RosettaNet4
Figure 13 e-SCM BPR full scale program roadmap
, a globally supported standards development organization, to be verified as an international standard. The significance of setting the international standard by leveraging the result of TSMC/ASE supply chain integration was two folds: one was to expedite the integration of the companies in the value chain and the other was to strengthen the leadership of these two company in the semiconductor industry.
To manage such a major cross-company BPR project, a systematic organization with
4RosettaNet, formed by major companies from various industries, provide the standardized data infrastructure for integrating business processes. RosettaNet, formed in 1998, is a globally supported standards development organization for collaborative commerce, mainly in the high-tech industry. The data standards now defined are widely adopted by companies to conduct inter-company SCM. RosettaNet has six established global councils:
computer and consumer electronics (CCE), electronic components (EC), logistics (LG), semiconductor manufacturing (SM), solution providers (SP), and telecommunications (TC). The
organization has more than 500 world-leading organizations joining and working to create, implement and promote open e-business standards and services.
strong leadership and effective governance is imperative. As shown in Figure 14, the
strong leadership and effective governance is imperative. As shown in Figure 14, the