Instructor Introduction

Frank Tsao, Vice President of the IoT Business Unit at Dingxin Digital Intelligence, has long been deeply involved in the fields of smart manufacturing, machine networking, and factory automation, possessing extensive practical experience in factories across Taiwan, mainland China, and Southeast Asia. He has assisted numerous manufacturing clients in the electronics, metal processing, and biotechnology industries in implementing machine networking, equipment health management, operational dashboards, and logistics automation solutions. His expertise lies in using IT+OT data integration, OEE optimization, and lean management to help companies overcome challenges such as labor shortages, difficulty in recruiting 3K workers, inconsistent workforce skills, and incomplete supply chains, achieving "less manpower, fewer bottlenecks, and visible benefits" through digitalization and automation to upgrade to smart factories.

The New Southbound Policy is no longer just about "cheap labor": why talk about automation?

When people mention the Southbound Policy, most still think of "cheap labor." But Cao Yongcheng points out that the conditions for this new wave of the Southbound Policy are completely different!

Labor costs continue to rise in Southeast Asia. With the flow of information, no one is willing to do 3K (dirty, dangerous, and hard) jobs, and even if higher salaries are offered, it may not be possible to recruit people. In addition, the supply chain in some areas is not yet complete, and the quality and stability of employees are still being established. The risk of "relying solely on people to support production capacity" is becoming increasingly greater.

Meanwhile, the rules of the global market are changing: product lifecycles are shortening, small-batch, high-variety production is becoming the norm, customers are fickle, and orders are changing frequently. Companies must accelerate production line changes, new product introductions, and quality stabilization to survive in the competition. In this context, automation is no longer a optional "bonus," but a "necessary requirement" tied to the New Southbound Policy.

But he also cautioned:

Automation requires a new mindset; it's not about simply moving machines to a new factory with old ideas.

To truly achieve a leapfrog development overseas, the first step is not to buy equipment, but to restart and check whether your data foundation is reliable .

Before automation, make the data real: moving from paper-based data and experience to digitalization.

Cao Yongcheng shared a memorable experience: In his early years, a colleague followed the company south to set up a factory in Vietnam, where he was responsible for production management. The Taiwan headquarters would adjust the KPI indicators every month, requiring the Vietnam factory to continuously improve.

My colleague initially thought very seriously about improvement measures, but found that the local factory manager seemed unconcerned about the new KPIs and did not make any changes; however, by the end of the month, the targets were always met on time.

The reason can be summed up in one sentence:

"The factory gives the boss whatever numbers he wants."

If numbers cannot reflect the true situation on the ground, then all management decisions made based on numbers are merely "looking good." Therefore, he emphasized that the first step is digitization —not for the sake of form, but to make the numbers real and traceable.

2-1 IT side: Integrate paper-based processes into the system

In most factories, orders, work orders, production orders, inspection sheets, inspection reports, and material requisition/return forms still rely heavily on paper documents and personal experience. This leads to several common problems:

By digitizing and eliminating paper, this information can be uniformly placed into the system, establishing the basic framework of information flow and providing a basis for subsequent analysis, tracking, and auditing.

2-2 OT End: Enabling Machines to Communicate with Sensors

However, IT alone is not enough. What truly determines production capacity and quality is the condition of the machines and equipment on-site.

In many factories, scales are still pointer-type, and measurements are recorded manually. Blurred numbers, illegible handwriting, and overlapping labels often make it impossible to determine which entry is correct. Such information cannot be used to calculate losses, estimate efficiency, or perform yield analysis.

Therefore, he proposed the direction of digitization for OT terminals:

At the same time, quality data must be collected: whether it is AOI automatic inspection, measurement instrument records, or manual inspection results, they are all "outcomes" of the production equipment.

Production equipment is the "cause," and quality data is the "effect." To improve, the cause-and-effect relationship must be integrated.

2-3 IT + OT: Understanding "A Day in the Life of a Machine"

When IT information such as orders/work orders/manufacturing orders is integrated with OT information such as machine/sensor/quality data, a complete "day of a machine" can be depicted: from powering on, dispatching work, receiving molds and fixtures, adjusting the machine, trial production, first article inspection, mass production, to loading, replenishing, handling jams, unloading, cleaning, and maintenance, data is recorded for each time period and status.

This allows companies to truly "see through" the machines—knowing when they are in production, when they are on standby, why they stop, the reason for the stoppage, and how much output is affected.

Only on this basis can discussing OEE (Overall Equipment Effectiveness) be meaningful. International benchmarks can reach 85%, but many factories cannot even accurately state their own actual OEE because they lack reliable underlying data.

Replacing post-event reviews with real-time reporting: Equipment health management and situation dashboards

Many factory improvement meetings are held at the end of the month, quarter, or year: analyzing reports, comparing figures, and questioning the reasons.

But Cao Yongcheng bluntly stated:

"Instead of reviewing data and investigating the root cause after the fact, it's better to notify immediately and take immediate action."

When a machine experiences an abnormal shutdown, a drop in production capacity, or an abnormal yield, if the system can push notifications in real time via a mobile app, tablet, or electronic dashboard, relevant personnel can immediately go to the site to handle the situation and prevent further losses. If the situation is not handled, the supervisor will also see the notification, creating a natural pressure and tracking mechanism.

Similarly, inspections, patrols, maintenance, and repairs also need to be digitized: traditional paper-based checklists make it difficult to confirm whether they have actually been performed; by taking photos and uploading them via mobile devices and filling in data in real time, the system can proactively remind users on a periodic basis and track whether the work has been completed, which helps improve the machine's reliability.

He shared a case study of a biotechnology production line:

By simply connecting the furnace tube machine to the network, along with mobile inspection and patrol, real-time reporting via the situation dashboard and APP, the production line utilization rate increased by 10% and production efficiency increased by 75% within just one month, resulting in a reduction of more than one million yuan in actual costs each year. Even the owner was amazed by the effects of digitalization.

This case highlights one point: often, improving OEE (Outcome Effectiveness) doesn't rely on larger machines, but rather on more immediate information and greater on-site transparency.

From intelligent information flow to automated logistics: Evolving eyes, hands, and feet together.

Automation only truly emerges after a solid data foundation is established. Cao Yongcheng likens factory automation to the "eyes, hands, and feet" of a person:

Dingxin, together with its partner Kaiyou Technology, provides a variety of automation solutions, from screw fastening and packaging lines to automatic batching and smart warehousing, all of which have been implemented in real-world applications.

But he also repeatedly emphasized that automation is not the goal, but a means.

A New Mindset for Automation: Don't Automate for the Sake of Automation

Before discussing automation, he put forward three premises: the purpose of a business is to be legal, sustainable, and profitable ; and automation investment must also return to these three points.

Many automation solutions assume that "all people will be replaced," but they do not delve into management issues and bottlenecks. They only see "labor shortages" or "unstable quality" and try to replace the entire production line with full automation, which often results in high costs, long payback periods, and even disruption to existing operations.

Cao Yongcheng pointed out several issues that need to be clarified first:

In many real-world cases, a combination of manual and automated work is actually more efficient: One workstation had six machines and previously employed four operators. After analysis, it was found that the first two workstations were heavy, hot, and dangerous, typical of the 3K model; the latter two workstations had a better environment, lighter workload, and were easier to staff.

If you pursue "unmanned" operation, you may need two robotic arms to replace four people, which will result in high investment costs and a long payback period. However, if you adopt "one robotic arm + one or two employees", focusing on replacing only 3K front-end workstations, you can save most of the manpower while retaining back-end flexibility, resulting in better investment returns.

He pointed out:

The most efficient factories are usually not fully automated or fully manual, but a hybrid model that "makes good use of both people and machines".

Furthermore, when evaluating the return on investment (ROI), one should not only look at "saving a few people," but also take into account the increased production capacity, shorter delivery times, management efficiency, and quality stability brought about by debottlenecking , in order to achieve the true overall benefits.

Don't lose at the starting line: When building a factory in the New Southbound Policy, you need to plan the energy pipeline, carbon fiber, and warehousing design.

Factories that have been operating on both sides of the Taiwan Strait for many years are often burdened by old power distribution and traffic flow systems. When they want to introduce energy management, carbon fiber and smart warehousing, they often have to pay high "restructuring costs". Building factories in the New Southbound Policy is a rare "blank slate opportunity".

6-1 Electricity and Carbon Footprint: Think Clearly About It From a Single Line Chart

In the future, carbon footprint and carbon inventory will inevitably become international standards. Many old factories have multiple layers of electrical distribution boxes, with main units and peripherals each having its own wiring. Over time, tracing which line belongs to which production line and which piece of equipment becomes extremely difficult.

If the power distribution architecture is planned and communicable meters are installed during the new factory phase according to the production line area, equipment type and carbon emission calculation requirements, a large amount of retrofitting costs can be saved when conducting energy consumption analysis and carbon footprint verification in the future.

6-2 Warehousing and Traffic Flow: Avoid turning a new factory into a "giant warehouse"

Industry surveys show that products are typically used for less than 5% of processing and manufacturing; the remaining 95% is spent on storage, waiting, and handling. If warehousing isn't planned well from the outset, common problems include:

In contrast, if the warehouse space is planned during the factory construction phase based on the height, aisle width, rack configuration, and the possibility of automated warehousing , and if the future needs to be considered for the introduction of AGVs, automated warehouses, or vision guidance, the cost of future adjustments can be significantly reduced.

To overtake on a curve in Southeast Asia, you must first see clearly, then act correctly.

For companies preparing to or already setting up factories in Southeast Asia, the question of "Should we automate?" is no longer the only one; the real key is:

In Cao Yongcheng's view, automation is not about showing off skills, but a long-term strategy based on data and oriented towards efficiency and effectiveness .

Building factories in the south is a marathon. If digitalization, automation, and management systems are integrated in the design from the starting point, companies will have the opportunity to truly achieve "going south while automating" in the Southeast Asian market and steadily overtake competitors.

[Service Inquiry Form] Let any idea turn into action!