Reading time ( words)
By James Scholler, MEC
In compressed product development cycles, traditional checks and balances become insurmountable constraints, driving non-value-added activity. A "freeze gate" lean concept leverages existing resources to meet a tight schedule without adding risk or impairing quality.
Product development and electronic manufacturing teams often face conflicting goals. There is a demand for leading-edge technology, faster product development cycles, and efficient cost-effective manufacturing processes. At the same time, teams are being asked to do more with fewer resources. Traditional product development and new product introduction (NPI) processes provide good checks and balances for work among cross-functional teams. However, in compressed product development cycles, those checks and balances can become insurmountable constraints rapidly, lengthening the cycle and driving non-value-added activity. This article uses a real-world example to illustrate how a traditional product development process evolved into a "freeze gate" process capable of leveraging existing resources to meet a compressed schedule, without adding risk or impairing overall product quality. The result is a process that delivers products faster with a smoother transition and incorporates lean concepts.
The product scope was an interface module that would incorporate remote network communications to provide control signals to a motor controller. Challenges included:
- •packaging must meet IP67 (able to experience a pressure wash down without performance irregularities);•product required multiple viewable indicators; •there were numerous M22 and M12 connectors; •product cost targets and timelines were set aggressively.
The electrical and mechanical challenges posed no overt challenges; however, the delivery date requirement was considered unachievable. The major schedule constraint was a well-thought-out ISO process with clearly defined stage gates (Figure 1).
Stage gates were developed to mitigate the risk of conflicting activity when operating in a cross-functional development team environment. However, working sequentially, they also have the net effect of shutting down the project while the activities required to clear the gate are completed and approved.
Figure 2. A shift is required to meet the tight time-to-market requirements. Deviation from standard workflow is needed.
To address this problem, the overall flow of the activities from the start of development to production delivery was evaluated. The flow was examined from a knowledge transfer approach. Product knowledge is limited initially; over time that knowledge grows. When an EMS provider is contracted, it is assumed that complete knowledge handoff occurs on Day 1. In reality, some knowledge transfer continues throughout the NPI phase and timing on key decisions can cut lead times.Next, the engineering team had to show the OEM what changes were required to meet the customer delivery date (Figure 2). Several manufacturing activities would need to start prior to design phase completion. One option was to shrink the knowledge transfer functions of each activity. The design phase, one of the longest functions, was a prime candidate. However, that meant eliminating features or enlarging the team to shorten the development. The OEM could not accept the loss of any features associated with this product and other design resources could not spare substantial time.
All key suppliers and engineers began to evaluate an appropriate approach to meet the delivery date. If certain decisions were made incrementally as the project developed, it would allow knowledge transfer to the other activities. Steps would start earlier and transfer knowledge in smaller packets, which facilitated better time management.
The key project stakeholders held a meeting to define key team members and their roles, as well as delineate the major risk factors in the program and the individuals responsible for solutions. The OEM understood the approach deviated from published ISO procedures. To manage risk, a list of the decisions made about the various features was kept with the appropriate approvals and the likely cost and time impact of changes to those decisions communicated.
Implementing the New ProcessHigh-risk development challenges were addressed first. For example, sketches on how to handle the sealing around the connectors were exchanged. Ideas were shared, such as using liquid compound filler to provide that sealing.
Another issue involved the best way to optimize component layout for manufacturability, while still providing for indicator viewing. The easiest approach was using leaded LEDs mounted to the top of the enclosure with the leads soldered to the PCB. However, this would have substantial complications in LED-to-case assembly and PCBA mounting. Light pipes from surface-mounted LEDs to the top of the enclosure represented a viable solution. The challenge was to provide a means to seal around the LEDs on the PCB and around the light pipes at the enclosure interface when the assembly was filled with the appropriate liquid sealing compound.
The various control cables and wiring, which connected to the main motor control board, presented a problem. Sketches were again exchanged that demonstrated ideas on how to best accomplish the task.
These early sketches were used to fix concepts to allow mechanical design to begin. A liquid compound for sealing around the connectors, the light-pipe approach for viewing the LEDs, and the rubber sealing means around the protrusion on the bottom of the enclosure were all fixed and OEM-approved.
The next step was tapping supplier expertise to help carry these early concepts to a defined process that would optimize manufacturability. One supplier had suggestions for decision timing that could compress the development process for enclosure injection molding tooling. The tooling supplier stated that once part size was fixed, they could place the order for the die. Once the part was designed with all steel-safe features, they could begin to cut the tool. When the final details of the enclosure were defined, they would then complete the die.
This approach was used in all aspects of development. The electronic hardware engineer could proceed without knowing all the details on the type/style of connectors. Early prototypes could be developed and tested to ensure that core functionality was understood. These prototypes looked little like the final configuration, but provided a reassurance early in the project that nothing was misunderstood. While enclosure and electronic design were progressing, software was concurrently developed with releases of core functionality occurring with the early prototypes. This affirmed that the unit's critical functionality was operational and any bugs were small in scope and easily resolved.
By making incremental decisions, those decisions were delayed until the latest possible time with the maximum accumulated product knowledge.
The EMS provider, a lean adopter that set up production requirements in a pull system with a kanban arrangement, was integral to this development. Early supply chain involvement also helped address long-lead-time component issues.
The same concepts applied in development stages were supported in the EMS NPI process. To complete the component supply chain early enough for manufacturing to meet production delivery needs, the bill of materials (BOM) was released in segments. This supplier discussion began while prototypes were undergoing acceptance testing. In development, the BOM often is largely complete at prototyping. There may be passive component changes required following prototype tests; however, these tend to have short lead times and be readily available. Most often, the major components are known early and are longer-lead-time items, such as microprocessors, specialized ICs, or displays. These components require more detailed arrangements with the supplier to support kanban arrangements, thus, longer negotiation timeframe. Consequently, the long-lead components on the BOM were fixed and OEM-approved at the time of the prototype.
The project moved along a timeline that would meet the OEM's requirement. Late in the development process, the OEM decided that an identification feature was critical and needed to be incorporated into the end product. The feature affected enclosure design, requiring an undercut in the part. The mold tool, which was underway, was designed as a simple two-cavity die without any need for lifters. If accounted for in the original tool design, the new identification feature would have modified the die to include lifters and would have required a different die base. To resolve this potential roadblock, the mold designer suggested a solution that kept the same die base, drastically reducing the cost/time impact. This process demonstrated the value of earlier risk management discussions.
ConclusionThe freeze gate development process breaks barriers to a free-flowing development and NPI process ? allowing the lean concept of "keep it flowing" to win out. The process of freezing focuses on product features, not documents. The development team along with key suppliers must be capable of defining what features need to be fixed and when. Process evolvement incorporates freeze gates in the development timeline. The freeze gates need to be along the critical path and have a scope that flexes with the scope of the project (Figure 3).