Zero Major Forklift Accidents Continued: The Full Story of IoT-Driven 'Safety Awareness Reform'
Reading this article will provide hints for solutions to challenges such as:
● Seeking a fundamental move to break through the stagnation of safety measures and on-site complacency where forklift accidents and near-misses occur repeatedly.
● Being burdened by data collection and aggregation tasks from drive recorders, leaving insufficient time for the on-site training and direct guidance that should be the primary focus.
● Failing to objectively grasp individual driving characteristics in an environment where many workers, including those from partner companies, are mixed together.
1. Summary
This article introduces a case study of strengthening forklift safety quality at a logistics center operated by a logistics subsidiary of a major electronics manufacturer group. A major accident occurred just four months after the center opened. Starting from the deep reflection that they were "not utilizing" the cloud-based forklift-specific drive recorders they had equipped, they implemented six measures to exhaust all functions. Furthermore, after achieving their goals, they evolved into an "activity revitalization phase" and continue to maintain zero major accidents to this day.
2. Background: Warehouse Forklift Accidents are Not 'Someone Else's Problem'
According to Ministry of Health, Labour and Welfare statistics, thousands of labor accidents involving forklifts occur every year, and fatal accidents are endless. It is known on-site that accident risks increase particularly under the following conditions:
･ Immediately after a new warehouse is built or significantly expanded, where workers are not accustomed to the environment.
･ When there is a large fluctuation in workload, causing a concentration of burden on specific workers.
･ When multiple employment types, such as partner companies and temporary staff, are mixed.
･ When managers are occupied with data aggregation and report creation, causing on-site guidance to fall behind.
While logistics centers are becoming larger and more automated,
forklifts remain machines operated by humans, and human error cannot be completely eliminated. That is precisely why the question is, "How quickly can we detect danger, and how reliably can we connect it to education and correction?"
3. Challenges on-site: A major accident occurred 4 months after opening
The stage for this case study is a logistics center operated by a logistics company of a major electronics manufacturer group, which mainly handles automotive equipment. This center, completed in August 2021, is a state-of-the-art facility that has introduced many labor-saving facilities such as high-rise automated racks and automated packing lines. On the other hand, it operates 36 forklifts and handles a wide range of tasks from parts receiving and shipping, JIT (Just-In-Time) supply, to domestic and international product shipment. The workers consist of staff from four partner companies, including our subsidiary, and from the beginning of the center's opening,
all 36 units were equipped with cloud-based forklift-specific drive recorders
. After the center began operations, they started safety quality improvement activities with the goal of one year without accidents, but in December 2021, just four months after operations began, a major accident occurred.
When trying to scoop up a load in the front with a long-fork forklift, the operator did not notice that the tips of the forks were caught on the cargo behind it and lifted the lift, causing the products on the upper level to fall. When checking the actual situation with data, there were two structural factors in the background.
Factor 1: Excessive workload on specific workers The person involved in the accident was responsible for unloading 50 trucks a day by themselves and was constantly in a state of "impatience."
Factor 2: Education had become a mere formality due to the confusion of the startup period Both managers and workers were busy with their tasks, and forklift safety education (confirmation of compliance with the 5 rules of safe driving) was not functioning sufficiently. When checking the footage before and after the accident using the "call-up function" of the cloud-based drive recorder, multiple rule violations against the 5 rules of safe driving established within the company were confirmed.
"Even though we had introduced drive recorders equipped with IoT technology, we were not effectively utilizing their functions. Any advanced tool is meaningless if it is not used."—This deep reflection led to the subsequent thorough strengthening of utilization.
4. Solution: 6 measures to exhaust 'all functions' of cloud-based drive recorders
▮ Why couldn't it be solved with conventional operations?
Cloud-based drive recorders were equipped with various functions such as immediate notification of dangerous driving, personal identification via RFID, and automatic aggregation via a dashboard. However, in the busyness of the startup period, a state of "notifications are received but put off" and "data is collected but aggregation is troublesome" continued, and it was not functioning for the original purpose of "preventing accidents before they happen." Based on the recognition that "even if there are tools, it is meaningless without design of how to use them and thorough operation," after the major accident, they launched a safety quality improvement project with all center members participating with the goal of
"zero major accidents in two years." While rotating PDCA twice, they implemented the following 6 measures in stages.
Measure 1: Thorough 'real-time immediate education' when dangerous driving occurs
When 3-sudden driving (sudden start, sudden braking, sudden turning) or impact is detected by the G-sensor, an email notification is sent to the manager in real-time. After the improvement, the operation was switched to one where the manager who received this notification immediately rushes to the site and conducts immediate education based on the 3-Gen principle (Genba/actual site, Genbutsu/actual object, Genjitsu/actual reality). The shift from "looking back later" to "educating right now" became the starting point for awareness reform.
Measure 2: Realization of 'personal identification x personal guidance' utilizing RFID
By holding an RFID card before boarding, it records "when, who, and which forklift was ridden." Since an alert continues to sound if the RFID card is not held up, it also functions to
prevent unauthorized boarding. Using random videos and call-up videos, they
analyze the driving characteristics and habits of each worker pinpointed, and multiple managers consider and provide optimal advice. Voices were raised from workers saying, "A sense of tension was born that we are always being watched."
Measure 3: Regularization of 'QY/KY training' using actual dangerous driving footage
Using individual videos as samples, they conducted Quality/Danger Prediction (QY/KY) training sessions once a month for all employees. Using the 4-round method (grasping the current situation -> pursuing the essence -> planning countermeasures -> setting goals), they made action declarations of "we will do this" in groups, aiming to
foster a culture where we follow the rules ourselves.
Measure 4: 'Work leveling' using operation rate data
To address 'overload on specific operators,' which was a factor in major accidents, we analyzed the driving time of each forklift using operational time data.
After visualizing through data that the workload of the individuals involved in accidents was abnormally high, we implemented two measures: A) Work reorganization (sharing tasks with operators in adjacent areas) and B) Establishing a support system (other operators provide support on high-load days).
Measure 5: Reduce 'data aggregation time' for managers and secure time for training
With the conventional SD card system, it took about 298 hours per year to create training materials, but the cloud-based dashboard function enabled automatic aggregation and data conversion, reducing material creation time to about 24 hours per year.
The 273.6 hours saved annually could all be dedicated to direct on-site training.
Measure 6: 'In-depth analysis of accident causes' using video and special practical training
As a result of retrieving and analyzing video from before the accidents, it was found that the individuals' weaknesses were 'insufficient confirmation of the safety space' and 'improper fork insertion.'
To overcome these, we repeatedly conducted special practical training to prevent recurrence.
This special training continues to be provided to new employees today.
5. Concrete Results: Dangerous driving reduced from 231 to 7 times per month, maintaining zero major accidents
The number of 'three sudden' driving incidents (sudden start, sudden stop, sudden turn) decreased continuously every half-year, and dangerous driving by all lift operators dropped sharply.
This improvement in figures is largely attributed not only to the effects of education and guidance but also to a 'change in awareness regarding alerts sounding during three sudden driving incidents.'
By having the system automatically detect dangerous driving, operators themselves began to take spontaneous safety actions to 'avoid triggering alerts'.
6. Evolution After Achieving Goals: Moving to the 'Activity Activation Phase'
The results of the improvement activities at one location were reported to the company's management, and the additional introduction of cloud-based dashcams to 21 locations and 286 vehicles nationwide was decided.
Quality control regulations were established at headquarters, and safety quality enhancement activities started on a nationwide scale.
Then, after confirming the continuation of zero major accidents in March 2024 and expanding nationwide, our center did not stop its safety quality activities.
Instead, we set 'how to prevent established activities from becoming mere formalities' as our next theme, evolving into an activity activation phase.
Initiative A: Introduction of an award system based on KPI management of 'three sudden' driving
To continuously activate overall activities, we introduced a unique award system that turned each operator's driving data into KPIs.
Specifically, we extract 'number of three sudden driving incidents (a)' and 'operating time (b)' on a monthly basis and calculate 'how many three sudden incidents occur in how many minutes (b÷a)' as a KPI index.
We calculate a comprehensive ranking by combining the operating time ranking and the KPI ranking, and award the top performers once every half-year.
The introduction of the award system created a culture of 'competing for safe driving with numbers,' succeeding in maintaining operators' safety awareness at a high level continuously.
We also set a target value (600 minutes/month or more) for the overall three sudden driving KPI and established a mechanism for the entire center to track the figures.
Initiative B: Improving communication through video check comments
Utilizing the comment section of the random video check sheet, we are promoting an initiative where managers add 'warnings, advice, and gratitude' when handing them to operators
. Although the content is simple, such as 'Please check XX a little more (warning),' 'It would be even better if you did YY (advice),' and 'Thank you as always (gratitude),'
two-way communication mediated by data is being born. Safety quality activities often tend to be one-way from managers to operators, but this initiative narrows the psychological distance with the field and leads to further improvement in safety awareness.
The insights from these 'activity activation phases' (KPI award system, comment sheets) are also being considered for sequential horizontal expansion nationwide.
7. Horizontal Expansion and Applicability: Effective for these companies and challenges
This approach is particularly effective for companies and logistics managers in the following situations:
- Operating many forklifts and unable to keep up with safety management
- Using SD card systems and spending a lot of time on data aggregation and utilization
- Wanting to grasp 'individual habits' that managers overlook during patrols
- Difficult to provide uniform training due to a mix of staff from multiple partner companies
The most important implication of this case is that 'simply introducing tools is meaningless'.
Despite the cloud-based dashcams being equipped from the start of operations, major accidents still occurred.
The 'determination to use them to the fullest' and 'organizational operational design' after introduction are the essence of improvement.
8. Summary: 3 Principles that turned 'IoT tools' into 'weapons for education'
We summarize the consistent thinking in this improvement case into three points.
① Maximize real-time capability ——
Bring the time lag from the occurrence of dangerous driving to education and correction close to zero.
The shift from 'looking back later' to 'educating right now' was the starting point for awareness reform.
② Identify individuals and optimize individually ——
Pinpoint 'whose, what kind of habit, and how to fix it' using RFID + video data.
It elicited spontaneous improvement actions from operators that cannot be obtained through uniform guidance.
③
Leave manager man-hours to the system and focus on work only humans can do —— Transfer data aggregation and report creation to the system and invest all the time created into on-site education.
Achieved 'labor saving in management' and 'strengthening of guidance' simultaneously.
And continuing to fight against the 'hollowing out of activities' even after achieving the goal is the key to long-term zero-accident continuity.
The award system based on KPIs and video check comments can be considered one answer to that mechanism.
At our company, we have accumulated know-how in 'organizing safety quality issues from multiple perspectives and solving them by combining tools and operational design'.
Why not consider logistics improvements that achieve both zero accidents and on-site education with us?
9. Inquiries
If you are thinking, 'I want to strengthen forklift safety management, but I don't know where to start,' 'I have introduced dashcams, but I am not utilizing them well,' or 'I want to optimize my logistics,' please feel free to contact us.
First, we accept consultations via our inquiry form.
📩 Inquiry Form: [
https://forms.cloud.microsoft/r/AhaEAjizXW]Questions in the comment section are also welcome. We would be happy to exchange information with others in logistics, manufacturing, wholesale, etc., who have similar concerns.
Furthermore, please see our official website for details on logistics solutions, services, networks, and technologies provided by MD Logis.
🏢 Website:[https://www.mdlogis.co.jp/]
<Related Articles>
▪Solving Traffic Line Congestion! Visualizing Warehouse Issues Using IE Methods
https://note.com/mdlogis_official/n/n44dccea041e1
▪Zero Visual Errors, 1 Worker Reduced—Improvement Techniques at a Logistics Center in Thailand
https://note.com/mdlogis_official/n/n8d484a1d0e68
▪57% Reduction in Work Man-hours—The Full Record of Parallel Robot Introduction in an Operating Warehouse
https://note.com/mdlogis_official/n/nbed831fca171
#Logistics #LogisticsManagement #SafetyManagement #Forklift #IoT #Dashcam #WarehouseImprovement #LogisticsDX #WorkStyleReform #LaborAccidentPrevention #NewLogisticsEfficiencyAct