【正文】 . The increase in productive moves causes the time spent on productive moves to go up from 62% to 66%, as shown in the bottom graph, Figure 13. Idle percentage decreased from 19% to 16%. The remaining idle time shows there is still room for improvement. Step 5B: faster quay cranes (and NO increased twin percentage) The (19902020) dual trolley quay cranes in the original scenario and that have been used up to now, are relatively landside hoist has an average cycle time of 99 seconds. With modern cranes cycle times of 63 seconds should be possible. The kinematics of the cranes in the model have been adjusted in Step 5B to be able to make cycles of 63 seconds. Expected effects: The quay cranes can make more cycles per hour and hence productivity should increase. Waiting times for LiftAGVs at the quay cranes should decrease since the cranes need less time per move, and hence can serve the next LiftAGV sooner. Results The quay crane productivity increases by 5 to 7 bx/hr, or 20%,as shown in Figure 14. Other effects that were observed after this adjustment: ? The quay crane status representing productive activity decreased from 90% to 65%. ? The liftAGV waiting and interchange times at quay cranes decreased from 220 to 100 seconds per box processed. ?The idle percentage of waterside RMGs decreased from 19% to 11%, and productivity increased from 62% to 73% (note: the differences do not even out because the landside RMG took over more unproductive work when the waterside productivity was increased) Step 6: all adjustments bined The final step is a parison between the start scenario and all adjustments described in the previous steps. We will see the overall impact on performance levels. Quay crane productivity has increased with bx/hr in the experiments with five vehicles per QC – or 68%! Remember that in Step 2, with the increased throughput, we already stated that QC productivity needed to go up to between 40 and 42 bx/hr and this goal has been achieved. The increased quay crane productivity is only possible with more efficient LiftAGVs and RMGs. Figure 16 shows that the LiftAGVs in the final scenario only need 7 minutes to plete one container move, while originally the AGVs needed 11 minutes. With the increased waterside productivities the stress on the yard has increased as well. The terminal throughput and according gate volume cause additional moves in the yard. The gate report shows that the RMGs are able to cope with this increased demand, because 460 truck moves have been handled and the truck service times are still acceptable, as shown in Figure 17. The increased demand on the yard is represented in the graph with RMG moves per stack module. In Step 6, the two RMGs in each stack module executed vessel boxes and gate boxes per hour, about 50% more than the original scenario. Meanwhile the number of housekeeping moves has been heavily reduced, as shown in Figure 18. This is not because there is less time, but because there is less need to do those moves. In the original scenario with dual RMGs, the RMGs often had to drop stackin containers as fast as possible to cope with local peak demands. Those containers needed to be transferred further away from the interchange areas later to make that space available again for use during new peaks. The twinRMGs didn‘t have that need, because they were able to deliver stackin containers to good slots immediately. The status chart of RMGs shows that the RMGs in both the standard and the final scenario are approaching their limits of activity, as shown in Figure 19. With less than 10% idle time there is too little flexibility to cope with local peaks in the yard. Conclusions In this paper we described a stepbystep approach to improve existing largely automated terminals to stateoftheart terminals and what each step can bring. Besides faster truck and vessel handling the described adjustments lead to a throughput increase of almost 50%. Actually adjusting existing terminals with the described changes is a costly and timeconsuming operation。我們 使 仿真 模型 集裝箱碼頭 處于特定的環(huán)境條件下，以此 來(lái)衡量 每 次 調(diào)整 對(duì)它的單獨(dú) 影響 。 這個(gè)中轉(zhuǎn)站 適用于每年完成 220 萬(wàn)標(biāo)準(zhǔn)箱的吞吐量 （標(biāo)準(zhǔn)箱不均衡系數(shù) ）； 這里 的轉(zhuǎn)運(yùn)不 超過(guò) 5%。 我們 對(duì)不同的調(diào)整進(jìn)行了 總結(jié) ，并描述了對(duì) 出 中轉(zhuǎn)站 生產(chǎn)力上的預(yù)期影響。這樣 就 可以提高 它的 性能。對(duì)于 長(zhǎng)時(shí)間 的 陸側(cè)服務(wù) ， 將會(huì)產(chǎn)生一種 負(fù)面 的 影響： 服務(wù) 時(shí)間超過(guò) 10 分鐘，這 就 意味著卡車(chē)不得不 在軌道式起重機(jī)轉(zhuǎn)讓地帶 的平均 等待 時(shí)間超過(guò) 10 分鐘， 直到那里 的集裝箱 被運(yùn)走 ！ 當(dāng)我們使用 兩臺(tái)軌道式起重機(jī) 時(shí) ， 卡車(chē) 的 服務(wù)時(shí)間大幅度減少， 以 一臺(tái) 軌道式起重機(jī)在陸側(cè)專(zhuān)用 。 門(mén)道通過(guò)率 提高到每小時(shí) 470 箱。 對(duì)于 陸側(cè)來(lái)說(shuō)，表現(xiàn) 出 了 更大的影響 。 第三步： 用可升降的 自動(dòng)導(dǎo)向車(chē) 代替 自動(dòng)導(dǎo)向車(chē) 在堆場(chǎng)中， 普通的 自動(dòng)導(dǎo)向車(chē) 需要一個(gè) ―抖動(dòng) ‖來(lái)與 軌道式起重機(jī) 起吊貨物進(jìn)行轉(zhuǎn)運(yùn) ，等待時(shí)間為 軌道式起重機(jī) 和自動(dòng)導(dǎo) 向 車(chē)兩部分 所產(chǎn)生總和 ，因?yàn)?對(duì) 幾乎每一個(gè)移動(dòng)， 它 們中的 一方 都 必須 等待另 一方 的到達(dá)。 這 在他們的 移動(dòng)路徑 中是一個(gè) 額外的移動(dòng)并花費(fèi) 了 額外 的 時(shí)間（ 每次 停留15–25 秒） 。 圖 8顯示了 在 自動(dòng)導(dǎo)向車(chē) 和 可升降的自動(dòng)導(dǎo)向車(chē)兩種情況下，轉(zhuǎn)運(yùn) 每個(gè)集裝箱 的持續(xù) 時(shí)間 。 新一代 的可升降的自動(dòng)導(dǎo)向車(chē) 能夠更快地 進(jìn)行直線運(yùn)動(dòng) ，曲線 運(yùn)動(dòng) 和減速 運(yùn)動(dòng) 。 在 第 4步 以 后， 在堆棧中的模塊 ，無(wú)論是在水上還是陸 側(cè)的軌道式起重機(jī) 都 會(huì) 有 19％的空閑時(shí)間。 它 們的空閑時(shí)間將會(huì)減少，生產(chǎn) 能 力 就會(huì)增加 。陸 側(cè) 的起重機(jī)的平均周期時(shí)間為 99 秒。 ?水上空閑的軌道式起重機(jī)的百分比從 19%下降到了 11%，生產(chǎn)率從 62%上升帶了 73%（注意： 因?yàn)楫?dāng)水上的生產(chǎn)率增加時(shí)，陸側(cè)的軌道式起重機(jī)承擔(dān)了更多非生產(chǎn)性的工作，所以差異性沒(méi)有超出 ） 步驟 6： 所有的調(diào)整相結(jié)合 最后一步是開(kāi)始場(chǎng)景與在前面所述的所有調(diào)整的一個(gè)對(duì)比。報(bào)告顯示，軌道式起重機(jī)能夠處理這個(gè)增加的貨運(yùn)量，因?yàn)?460卡車(chē)是能夠處理這些搬運(yùn)的，卡車(chē)的運(yùn)行時(shí)間任是可以接受的，如圖 17所示。 軌道式起重機(jī)的狀態(tài)圖表顯示，在標(biāo)準(zhǔn)場(chǎng)景和最后一個(gè)場(chǎng)景中， 軌道式起重機(jī)都接近達(dá)到了它們的極限生產(chǎn)率，如圖 19所示。 此外，該研究證明，盡管與目前的經(jīng)驗(yàn)相比，模擬 出 的結(jié)果看上去似乎太高，但我們的步伐正與 先進(jìn) 的技術(shù)靠近 — 在相同類(lèi)型的仿真模型中 ，這個(gè)模型 具有廣泛的代表性 — 未來(lái)先進(jìn)的技術(shù)是 簡(jiǎn)潔 的而且在 很大程度上 都是 可行的。他在1996年創(chuàng)立了