The 1970s played a pivotal role in the history of trade, shipping, and manufacturing. In this decade, Toyota revolutionized the manufacturing world with the wide adoption of its Just-in-Time (JIT) production planning process. In essence, that represents operating with low inventory levels where raw materials, goods, or even labor are scheduled to arrive or be replenished exactly when, or shortly before, they are needed in production. Having the system, Toyota could solve some of the problems associated with keeping high-inventory levels, free up a significant amount of capital, and reduce the opportunity costs. It could also minimize storage, service/maintenance, and inventory risk costs.
Nowadays, JIT and its equivalents are the industry standards in the automotive and are also universally adopted in the wider manufacturing and service verticals. However, if automotive or manufacturing sectors can afford to wait for an order to be placed and then produce, in many other business areas, in order to adopt JIT, highly accurate Demand Forecasting must be in place.
An obvious example of such an area is the container shipping industry, which in the 1970s has also experienced a renaissance with the wide adoption of containerization. However, improvident internal processes and trade imbalances led to huge inefficiencies. Empty containers had to be repositioned from big consumer centers such as Europe and the US to big manufacturing powerhouses like China and Southeast Asia. To be able to satisfy demand at any given time and location, and compensate for the long repositioning cycles, ocean shipping carriers keep excess container inventory. Traditionally, inland empty containers account for around 30-40% of the total container shipping fleet.
This contrary to the JIT way of working is called Just-in-Case (JIC) – it represents an inventory-management method whereby inventory, goods, and materials are always on hand, to be at the right place when needed and to ensure that customer orders are always satisfied. However, the estimates for the safety stocks or inventory levels in the ocean shipping industry are often not data-driven and based on the gut feeling and experience of local agents. They are often highly incentivized to constantly keep a larger safety stock in order to be able to satisfy all potential customer orders. This incentive goes into conflict with the strive for efficiency of the overall organization.
The contrary scenario in which the central organization dictates the local safety stocks also exists. After the collapse of Hanjin in 2016 (which was among the world’s top 10 carriers back then) and the following turmoil, some companies have significantly slashed their fleets. However, that created the opposite out-of-stock problem, where companies are struggling to satisfy customer demand.
Regardless of whether out-of-stock or overstocking, the problem could be very costly. According to Boston Consulting Group (BCG), the cost for this empty container logistics (incl. repositioning, storage, handling, etc.) accounts for about 5-8% of operating budgets and is estimated to cost the industry around $20bln annually. Adding to that maintenance & repair, the costs can easily reach 12% of a shipping line’s operating costs. According to the CEO of the 6th largest ocean shipping line globally, Jeremy Nixon, dealing with empty inventory, is roughly a billion-dollar problem for Ocean Network Express (ONE).
The logical question here is how Ocean Shipping Carriers can switch from JIC to JIT to improve their operations and P&L. Contrary to the common understanding, JIT in ocean shipping should be easier to achieve than in manufacturing. Similarly, as in manufacturing, its implementation goes through building lean and extremely efficient supply chains. However, in the ocean shipping industry companies should not fully rely on external parties to always supply them on time with the needed goods and materials. Rather, ocean shipping carriers can rely solely on themselves, or to be more precise, on the utilization and accuracy of their own data.
Step by Step towards Just-in-Time Shipping
As discussed above, a prerequisite to adopting JIT is to precisely know the future demand. Currently, the expected demand of smaller ocean shipping carriers and the safety stocks tied to it are primarily calculated by sales and local agents based on their customer interactions, experience, and gut feeling. The larger and more technologically advanced ocean shipping carriers rely on statistical modeling of their own data in order to detect trends and seasonality and to forecast demand. However, logistics data often falls below certain quality standards and, if not properly preprocessed, further data modeling may lead to the wrong conclusions.
Thus, machine learning algorithms should be deployed to increase and improve data usability and ensure high forecasting accuracy. However, the first immediate effect of this action is achieving full Inventory, State & Cost visibility across the entire network. On top of that, external factors that affect volumes (holidays, PMIs, economic indicators, etc.) should be also incorporated into the forecast to ensure even higher forecasting accuracy and robustness. This step is crucial for finding the right dynamic safety stocks at any given time per asset type, grade, and location.
To further optimize their fleet, companies can deploy smart optimization algorithms that take into consideration all relevant costs (storage, repositioning, stevedoring, handling, maintenance, etc.), sailing schedules, and bookings curves among others. The combination of demand forecasting and this cost-based optimization provides the possibility to make the data-driven assessment whether repositioning is meaningful or not, from which place to resupply (both locally & globally), whether to grade, and at which location storage or maintenance should take place. The ease and speed to resupply should be also taken into consideration when conducting the risk assessment. Integration or collaboration with short-term leasing vendors could further increase efficiency.
Arguably the highest short-term impact of JIT on a shipping company P&L is the possibility to minimize the total fleet size with up to 12% without hurting revenues and service levels.
If all of these elements are applied, ocean shipping companies should be able to build the highly efficient supply chain required to reap the benefits of JIT. Transmetrics’ customer implementations show that JIT could help shipping companies cut their storage costs by up to 30%. Additionally, empty container logistics costs could be optimized with further 10%. Arguably the highest short-term impact on a shipping company P&L is the possibility to minimize the total fleet size with up to 12% without hurting revenues and service levels. In fact, Transmetrics’ JIT system could also indicate where container shortages are likely to appear, weeks in advance. This even helps companies mitigate that and gain additional revenues by meeting demand where and when it occurs.
Despite the recent stable financial results shown by some of the ocean shipping companies, there is no doubt that 2020 has brought new challenges for the container shipping sector. In addition to the existing US-China disputes and already deteriorating trade, the COVID-19 pandemic causes further pains to the industry. In the most apocalyptic forecasts, some analysts say that the pandemic may lead to the “largest decline in shipping volumes in living memory”. All of this decreases profitability and puts additional pressure on operations, forcing shipping companies to strengthen their efforts to increase efficiency. Thus, there is arguably no better time for ocean shipping players to act and adopt Just-in-Time.
It took a while for Toyota to perfect the system, but it is fair to say that JIT helped the company build a competitive advantage over its rivals and revolutionized the automotive & manufacturing world. Now it is time to act for the container shipping lines, as all means for implementing Just-in-Time Shipping are present.
The Transmetrics project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 945610.