The Box: How the Shipping Container Made the World Smaller and the World Economy Bigger (11 page)

Matson, established in 1882, had been a loosely managed, family-dominated company that grew from a single ship in Hawaii into a transportation conglomerate. It owned California oil wells, oil tankers, and tanks in the Hawaiian Islands to store the oil. It owned passenger ships and built hotels on Waikiki Beach to attract passengers. It owned Hawaiian sugar plantations and the ships to carry sugar to the mainland. For a few years after World War II, it even owned an airline. None of this made much money, and the company’s underlying problem was that many of its big shareholders didn’t want it to make much money. The board of directors included representatives of major Hawaiian sugar and pineapple growers whose main interest was a cheap way to get their products to market. Whether the shipping service made a profit was almost incidental.
⁹

Things began to change in 1947, when the Matson family convinced veteran steamship executive John E. Cushing to postpone his planned retirement and serve for three years as president. Cushing put the company on a budget for the first time and took a serious interest in addressing dismally low productivity. In 1948, Matson installed a revolutionary mechanized system to ship sugar to the mainland in bulk rather than in hundred-pound bags. Bulk sugar had required large investments—huge bins at the Hawaii end to hold the raw sugar, a special fleet of trucks to carry the sugar from mills to the pier, conveyors to move the sugar from the trucks to the top of the bin, and more conveyors to recirculate it within the bins, so the sticky substance would not solidify in place. These outlays had brought vastly lower costs. Sugar had given Matson a feel for what automation could achieve. Shortly after Cushing’s departure, the company decided to look into mechanizing the handling of the general cargo it carried between the West Coast and Hawaii.
¹⁰

Matson moved deliberately. Pan-Atlantic, under McLean’s control, was a scrappy upstart building a brand-new business, and it risked little by acting quickly. Matson had no such haste; it had a large existing business to protect, and its directors were tight with the purse strings. After commissioning outside studies for two years—the same two years it took Malcom McLean to move from a concept to a functioning business—Matson created an in-house research department in 1956. The man recruited to run it was Foster Weldon, a geophysicist most recently involved in developing the Polaris nuclear submarine.

The contrast with Pan-Atlantic could not have been more stark. McLean’s engineers, people like Keith Tantlinger and Robert Campbell, were no intellectual slouches, but they had worked in industry, not academia, and they were well advised not to flaunt their pedigrees in public. Weldon was a professor at the prestigious Johns Hopkins University in Baltimore and a well-known figure in the new science of operations research, the study of efficient ways to manage complex systems. Pan-Atlantic’s initial technology had been designed on the fly, using obsolete tanker ships, shipbuilding cranes, and containers whose length was determined by the size of the tankers, on the assumption that it could all be improved once the business was up and running. Weldon found this catch-as-catch-can strategy bewildering. “All transportation companies have their own pet theories on the detailed equipment requirements comprising a ‘best’ container system, but there are no quantitative data relating even such gross characteristics as container size to the economics of a total transportation operation,” he wrote pointedly. His goal, as he defined it, was to develop good data and use them to find the
optimal
way for Matson to embark upon container shipping.
¹¹

Weldon quickly came upon the issues that would shape Matson’s approach. About half of the company’s general cargo was suitable for shipment in containers, but the flow was out of balance: for every ton the company shipped from Hawaii to the mainland, it shipped three tons from the mainland to Hawaii. Revenues from the westbound run would need to cover the cost of returning large numbers of empty containers west to east. Even worse, much of Matson’s business came from food processors in California sending small loads to mom-and-pop grocery stores in the islands. Matson would need to consolidate these small shipments to fill whole containers in California, and would then have to open the containers in Honolulu and parcel out the loads for various destinations. This would make container shipping expensive. On the other side of the equation, though, Weldon found that by eliminating the need to transfer individual pieces of cargo from trucks to ships and back again, containers would eliminate almost half the cost of Matson’s existing business. “[T]his cost has increased steadily in the past and will continue to do so indefinitely as long as the operation remains a manual one,” he concluded. “There is certainly no indication of a change in the current trend of spiraling longshore wages with no corresponding increase in labor productivity.” Given the urgent need to automate, Weldon conceived of a way to make the container work: if Matson could load those small shipments into containers in route-sequence order, delivery trucks could collect the containers in Honolulu and proceed immediately on their routes. The goods for each store would be handled only when the truck arrived there, making containerization on the Hawaii run economically viable.
¹²

Given that containers made sense, how big should they be? Weldon’s analysis pointed out that the smaller the size, the greater the number of loads that would fill entire containers going directly from shipper to recipient, with no reloading. On the other hand, two 10-foot containers would take twice as long to load on a ship as one 20- foot box, making poor use of the company’s investment in cranes and ships. After analyzing thousands of Matson shipments by computer—a task that in 1956 required feeding in thousands of punch cards—Weldon’s researchers concluded that vans of 20 to 25 feet would be most efficient in the Hawaii trade: larger containers would travel with too much empty space, while containers shorter than 20 feet would require too much loading time. They recommended that Matson start out by carrying containers on deck, as Pan-Atlantic had, with conventional breakbulk cargo in the holds. By converting six of its fifteen C-3 cargo ships to carry containers on deck, Matson would be able to offer weekly container service between Honolulu and both Los Angeles and San Francisco. Weldon found that this arrangement would be profitable even if the container business stayed small. If the business grew, the company could convert additional ships to carry only containers. Containerization, he concluded, “would appear to present the fortunate circumstance of a promising initial course of action offering the option of going as far as desired and stopping at any point that prudent planning dictates.”
¹³

Matson management accepted Weldon’s recommendations in early 1957. Leslie Harlander, a newly minted naval architect, was put in charge of the engineering. Harlander was told to hire a staff and begin detailed planning for every aspect of a container operation. He was given clear guidance to be careful about money. Every choice had to be justified based on whether it offered a higher return on investment than the alternatives.
¹⁴

Harlander and his brother Don, an engineer who specialized in cranes, began to lay out their requirements for cranes in July 1957. In October, they went to Houston to observe the first arrival of Pan-Atlantic’s newly rebuilt
Gateway City.
The
Gateway City
was a C-2 ship, slightly smaller and slower than the World War II-vintage C-3s in Matson’s fleet, and it was equipped with Sea-Land’s two novel shipboard cranes. With both cranes working, the
Gateway City’s
turnaround time was no longer than that of the much smaller
Ideal-X.
As the Harlanders saw firsthand, though, the shipboard cranes had shortcomings. Pan-Atlantic’s two crane drivers each sat high above the deck facing two colored lights. A green light told one driver that he could move the crane trolley over the side of the ship to deposit a container on the dock, while a red light told the other driver to wait. If both cranes accidentally dangled forty-thousand-pound containers over the side at the same time, the unbalanced weight could capsize the vessel. Matson, with plans to serve only a small number of large ports rather than many small ones, had no need to put up with this risk. The first big decision was an easy one: land-based cranes were the way to go.
¹⁵

These would not be leftover cranes adapted from some other use like the cut-down shipyard cranes Pan-Atlantic had pressed into service in 1956. The original Pan-Atlantic cranes were revolvers, known in a shipping trade as “whirleys.” They did well enough at picking up a container from the deck of a ship and swinging it in an arc toward the dock, but their design made it difficult to lower the container precisely atop a trailer chassis, which slowed down the entire operation. Matson’s cranes were designed from scratch, with a requirement that they be able to unload an incoming container and load an outgoing box within five minutes—a cycle two minutes shorter than that of Pan-Atlantic’s first cranes. The Matson cranes were to have booms that stretched ninety-five feet from the dock, more than enough to span the entire width of the ships in Matson’s fleet. The operator would control a trolley to move the lifting beam out over the ship, lower the lifting beam to pick up a container, hoist the container, and then travel toward the dock at speeds up to 410 feet per minute. At high speed, these movements would have left each container swinging from the long hoist cables, far above the deck. Les Harlander designed a special lifting spreader to solve the swing problem, testing its feasibility by building a model with his son’s Erector Set over Christmas of 1957.
¹⁶

Weldon’s work had concluded by recommending a container 20 to 25 feet long. Harlander had the job of getting a design developed. In late 1957, Matson engaged Trailmobile, a manufacturer of truck trailers, to build two prototype containers and two chassis. Another contractor constructed two lifting spreaders and a steel frame that would simulate a container cell within a ship. Months of testing followed. Gauges to measure strain were attached to the equipment, and the stresses were established as containers of various weights and densities were lowered into the cell, lifted out again, and placed on the chassis. The test cell was set at various angles to determine just how much clearance was needed between the containers and the vertical angle bars that formed the corners of the cell. Loaded boxes were stacked to measure the pressures on the bottom container, and lift trucks were run inside the containers to measure the strain on the floors.

When the results were in, Harlander’s team decided that the most economical size for Matson was
feet high and 24 feet long, 11 feet shorter than Pan-Atlantic’s containers. The specifications took into account Weldon’s finding that each pound of weight saved was worth 20 cents, each additional cubic foot inside the container worth $20. To improve structural integrity, the roof would be a single sheet riveted in place rather than several panels attached with sheet metal screws, the design Trailmobile used for highway trailers. Steel corner posts would have to be able to support 120,000 pounds—the weight of several stacked containers, and much more than the posts in Pan-Atlantic’s first containers could support. The doors, two layers of aluminum with stiffeners between, were designed to dovetail rather than to meet in a straight line, to withstand twisting pressure due to a ship’s rolling in a heavy sea. The floor would be Douglas fir with tongue-and-groove joints. Special attachments to make the containers compatible with specific cranes and forklifts were ruled out on grounds of cost. “It takes very little in the way of extra features to add, say, $200 to the cost of a container,” Harlander commented. “There would be a marked change in the total profit picture if the equipment costs were, say, 10 percent higher than they need to be to do the job satisfactorily.”
¹⁷

Early in 1958, as McLean was preparing to open Pan-Atlantic’s new route to Puerto Rico, the Pacific Coast Engineering Company (PACECO), the lowest of eleven bidders, won the contract to build Matson’s first crane. PACECO was not comfortable with the unusual design, and it declared that it would not be responsible for swinging containers, problems with the trolley, or difficulties working as fast as Matson specified. Harlander agreed that Matson would take responsibility for the design, and PACECO began work on an A-shaped monstrosity rising 113 feet from the dock, with legs 34 feet apart so that two trucks or two railcars could pass beneath the crane. Trailmobile built 600 containers and 400 chassis to Matson’s specifications. Matson developed a lashing system so that containers could be stacked up to five-high on deck, depending on their weight, without risk of damage at sea.
¹⁸

Meanwhile, Weldon’s research department pursued its quest for optimality by investigating the most efficient way to use Matson’s fleet. Renting time on an IBM 704 computer at several hundred dollars a minute, the researchers built a fully fledged simulation model of the business, incorporating data on volume and costs for more than three hundred commodities at every port the company served at every time of year. Then they added in data on port labor costs, the current utilization of docks and cranes, and the load aboard each ship, to provide real-time answers to practical questions: Should a big Hawaii-bound ship call at Hilo and Lanai, or should it transfer its cargo to a feeder ship at Honolulu? What time of day should a vessel depart Honolulu so as to minimize total costs of delivering a load of pineapple to Oakland? Such simulations were new in the 1950s and had never been used in the shipping industry.
¹⁹