Slips, Hours and Delivery Dates at Kuat Drive Yards: A Records Study of Capital-Ship Construction Before and After the Military Creation Act, 30–21 BBY
Abstract
The Military Creation Act of 22 BBY was followed by a rapid increase in orders for capital ships at Kuat Drive Yards, but whether the yard met the surge by building faster or by bringing more slips into service has not been examined against its own ledgers. We compared the eight standard years from 30 to 22 BBY with the first 420 standard days after the Senate's grant of emergency powers, using the Slip Office's ten-day muster register and the contract ledger. Capital-ship contracts rose from 41 (5.1 a year) to 27 in the window (23.5 a year), a rate ratio of 4.6 (95% CI 2.8–7.4). Slips in service averaged 11.6 before and 21.7 after, and capital-ship occupancy of available slip-days rose from 62.0% to 88.0%. On a logarithmic scale, 64% of the growth in occupied capacity came from added slips and 36% from fuller use. Mean build time for delivered hulls was 604 days before and 574 days after, a difference of −30 days (95% CI −85 to +25), and mean delivery slippage widened from 47 to 83 days (difference +36, 95% CI −6 to +78). Civil hull contracts fell from 5.8 to 2.6 a year (rate ratio 0.45, 95% CI 0.14–1.46). The window is short, only nine hulls were delivered in it, and all of them were laid down before the Act. We report the directions and decline to call them findings.
1. Introduction
The Slip Office of Kuat Drive Yards has kept a muster register of its building slips, in its present form since 30 BBY, and its contract ledger records for every hull a quoted build allowance, a keel-laying date and a delivery date. Both series are internal documents of the yard, and earlier studies of the Republic's naval procurement drew instead on the Senate's published estimates (Pasquin, 24 BBY), which record what was voted and say nothing of what a slip could hold, and on staff surveys of delivery performance that relied on the buyer's receipts (Quarn and Rennick, 22 BBY).
Here the yard's own records serve to look at one episode. In 22 BBY the Senate granted the Supreme Chancellor emergency powers, and the Military Creation Act became operative under that grant. Its reading as a broad standing authority with no term and a thin reporting duty is set out elsewhere (Delassar et al., 22 BBY); we rely on that analysis for the Act's authority and for nothing about procurement. What followed at the yard is the subject here. The Republic fleet became a fighting service (Rask, 21 BBY), orders for capital ships multiplied, and the question for a yard is always the same one. A surge in demand can be met by building each hull faster, by putting more hulls on more slips at once, or by some mixture, and each answer has a different cost for the customers who are not the navy.
Slip occupancy, build time and delivery slippage are compared across the Act, and civil hull contracts are counted on the same ledgers; this paper does both. The vantage is 21 BBY. The Clone Wars are under way, the Act remains in force, and the ledgers close at the last muster before this paper was drafted. Nothing is claimed about how the conflict will develop or about the yard's later output.
The post-Act window is brief. We state this at the outset because it governs every claim below.
2. Methods
The Slip Office musters every slip at ten-day intervals, entering whether it is in service, empty or occupied, and by which kind of hull. The contract ledger lists each contract with its date, the buyer class (Republic fleet or civil), the quoted build allowance in standard days, the keel-laying date and the delivery date where one exists. Both series were read from the yard's own copies. We work at the yard, and the consequences of that are discussed under limitations. Before the Act the Republic class covers hulls ordered for the Judicial Forces and the Republic's small standing squadrons; member-world orders are entered as civil only where they were built to civil specification. The counts are published with the clearance of the Republic fleet's procurement office, which reviewed the draft; no hull is identified.
Data before the Act run from 30 BBY, when the muster register in its present form begins, to the Senate's grant in 22 BBY and cover 8.0 standard years or 292 musters. The post-Act window opens on the day of the grant and runs for 420 standard days, or 42 musters, to the last muster before drafting. Contracts are assigned to the period in which they were signed, and hulls are assigned to the period in which they were delivered, which is why the delivered hulls in each period are not the hulls whose contracts were signed in it. The window contains 1.15 standard years in all, a standard year being 365 standard days, as the Slip Office reckons it.
Slip occupancy, the measure of yard load proposed by Oswin (23 BBY), is the share of available slip-days (slips in service multiplied by days) on which a slip held a hull of the stated kind. Build time is the interval in standard days from keel-laying to delivery. Delivery slippage is the interval from the delivery date written in the contract to the date of delivery; a negative value is an early delivery, and none occurred. Civil hull contracts are contracts for hulls built to civil specification, chiefly for freight and passenger carriage, and no particular design is meant. Contract and delivery rates are counts divided by years of exposure. Their ratios are given with intervals from the normal approximation on the logarithm of the ratio, which is adequate for the capital-ship counts and poor for the three civil contracts in the window. Means of build time, slippage and quoted allowance are compared between periods with an unequal-variance t procedure, using the degrees of freedom that follow from the variances. Six between-period contrasts are reported without correction for multiplicity; their p values describe the size of each gap against the scatter and are not tests of hypotheses. The growth in occupied capacity is split into a part due to more slips and a part due to fuller use by taking the logarithm of each ratio; the shares are shares of log growth, and they depend on that choice of scale.
The pre-Act period holds 41 capital-ship contracts and 46 civil contracts. Table 2 follows the capital hulls through the register. When the register opens in 30 BBY, 4 capital hulls from earlier contracts are on the slips. The 30 hulls delivered before the grant are those 4 and 26 of the 41 contracts. Of the other 15 contracts, 11 were on the slips on the day of the grant and 4 had been signed but not yet laid down. Nine of the 11 on the slips were delivered inside the window, and two were still building when the ledgers closed. The 4 contracts not yet laid down at the grant were laid down in the window, and so were 18 of the window's 27 new capital contracts, 22 in all, the first within 20 days of the grant and the mean at about day 164; the other 9 had not been laid down by the last muster. None of the window's own contracts had produced a delivery by then. The delivered samples are therefore 30 hulls before and 9 after, and the 24 capital hulls on the slips at the last muster are the 2 old ones and 22 laid down in the window. Carried through the flow, the hulls reproduce the occupancy figures below to within the rounding of the mean slip count.
3. Results
Orders rose sharply (Table 1). The ledger shows 41 capital-ship contracts in the eight pre-Act years, 5.1 a year, and 27 in the 420-day window, 23.5 a year, a rate ratio of 4.6 (95% CI 2.8–7.4). Civil hull contracts moved the other way: 46 before, 5.8 a year, against 3 in the window, 2.6 a year, a rate ratio of 0.45 (95% CI 0.14–1.46; p = .18). The interval for the civil ratio includes 1, and while its direction is clear in the counts, the count of three leaves the size of the fall uncertain.
Slips grew in number. The yard had 12 slips in service on the day of the grant and 27 at the last muster, a net addition of 15, which the register attributes to 6 reactivated slips and 9 newly commissioned ones. Averaged over each period, slips in service numbered 11.6 before and 21.7 in the window, a ratio of 1.87. Occupancy by capital hulls rose from 62.0% of available slip-days to 88.0%, a ratio of 1.42. Civil hulls held 12.0% of available slip-days before and 5.5% in the window, so that the share of occupied slip-days held by civil hulls fell from 16.2% to 5.9%. Total occupancy rose from 74.0% to 93.5% as capital and civil hulls together filled the larger yard.
Occupied capital capacity, which is slips in service multiplied by capital occupancy, was 7.2 slip-years a year before and 19.1 in the window, a ratio of 2.66. On the logarithmic scale, 64% of this growth is attributable to the larger number of slips and 36% to fuller use of them. The split would shift if the ratios were combined in another way, but under any ordering the added slips are the larger part.
Build time barely moved. The 30 hulls delivered before the grant averaged 604 days from keel-laying to delivery (SD 72); the 9 delivered in the window averaged 574 (SD 66). The difference of −30 days has a 95% interval from −85 to +25 (p = .26), and the ratio of means is 0.95. The direction is towards faster building, and the interval takes in no change at all and a lengthening of nearly a month.
Delivery slippage widened. Delivered hulls were late against their contract dates by a mean of 47 days before the grant (SD 49) and 83 days in the window (SD 52), a difference of +36 days (95% CI −6 to +78; p = .089). All nine hulls delivered in the window were laid down and contracted before the Act, so their delivery dates were fixed under the earlier regime. For contracts signed in the window, the quoted allowance was longer: a mean of 631 days (SD 66, n = 27) against 556 (SD 58, n = 41), a difference of +75 days (95% CI +44 to +106).
The yard also revised the promises it had made to civil buyers. Seven civil contracts were in force on the day of the grant, most not yet laid down, and each had its delivery date moved later; the mean extension was 112 days (95% CI 71–153 from the t distribution on 6 degrees of freedom; SD 44).
Delivered hulls per year rose from 3.75 to 7.8, a rate ratio of 2.09 (95% CI 0.99–4.39; p = .053). This ratio cannot be taken for a rise in the yard's output rate. Eleven hulls were on the slips on the day of the grant, and all nine of the window's deliveries came from them.
The window truncates the build-time comparison. Hulls laid down in it could not have been delivered unless built in under 420 days, and none were. The two hulls from the original eleven that were still on the slips at the close had been building for 640 and 655 days. If both are entered as delivered on the last day of the window, the post-Act mean build time becomes 587 days and the difference from 604 shrinks to −17 days. The true figure for the two hulls lies above that bound, so the observed direction towards faster building is partly an artefact of who has been delivered so far. This is the bias that Prell (22 BBY) describes for completion-time records, met here with a crude bound and not with a method for censored data.
4. Discussion
The pattern in the ledgers points one way. In the first 420 days after the Act, the added slips account for the larger share of the growth in occupied capacity, and no clear speeding of the building itself appears; whether the slips, and not faster building, carried the surge is a direction these data cannot yet settle. The build-time difference is small beside its scatter, and part of it may be an artefact of truncation, as the bound given in the Results indicates.
A widening of slippage is consistent with a yard that has taken on more than it can finish on time, though the interval on the nine delivered hulls reaches a small improvement as well as a large lengthening. The longer allowance quoted on new contracts is firmer. A mean quote 75 days longer is the yard's own statement of what it expected, made at signing, and it appears to be in line with a yard that regarded the added slips as insufficient on their own. Quoted allowances carry no information about what was delivered, and they may reflect a change in the mix of hulls contracted as much as a change in pace.
For civil buyers the direction is unfavourable. Their contracts were fewer, they held a smaller share of occupied slip-days, and the seven in force were pushed back by 112 days on average. The fall in new civil contracts rests on three events and cannot be called a measured decline. It is also uncertain whether buyers stayed away because the yard discouraged them, or because they expected the delays that the extension of existing dates then showed.
The results bear on one question and not on others. They bear on how the first part of the surge was absorbed. They do not say whether the yard will reach a rate of delivery that matches its occupancy, because the hulls laid down in the window have not yet been delivered, and the delivered-hull rate of 7.8 a year reflects a backlog that was already on the slips. A longer record, covering the delivery of the hulls now building, might show whether build times lengthen under heavier occupancy or hold at their former level, and until then the shorter build time seen here is only a possibility that the data fail to exclude.
Whether the Act's design mattered to the yard belongs to the analysis of the Act cited above. Nothing in the ledgers shows that the Act's lack of a term or a review bore on the yard's decisions, and we draw no inference about its design from them.
5. Limitations
The window is 420 days long and contains nine delivered hulls. Every statement about build time and slippage after the Act rests on those nine, all laid down before the Act, so that the comparison is between hulls built under the earlier regime for part of their time and hulls built under it entirely. The hulls that will reveal how the yard builds under the surge are still on the slips.
Records of this kind are right-truncated. The delivered sample is drawn from the hulls that have finished, and the unfinished ones are the longer builds. The bound of 587 days uses the elapsed times of the two unfinished hulls from the original eleven, and the 27 hulls contracted in the window are not represented at all in the build-time sample. Intervals computed from truncated records are probably too short, and the bound answers Prell's warning (22 BBY) only crudely.
Employment at the yard whose ledgers are used bears on the whole study. The registers were read in the yard's copies and not checked against the buyer's receipts, and a reader without access cannot repeat the count. The classification of a hull as capital or civil follows the ledger's buyer class, which the ledger does not define beyond the convention stated in the Methods. The attribution of the 15 added slips to reactivation and new commissioning is the register's own, and we have not inspected the slips.
Intervals for the rate ratios use a normal approximation that is weak for the civil count of three, and the six contrasts are uncorrected for multiplicity. The decomposition into slips and occupancy depends on the logarithmic scale. Finally, a before-and-after comparison cannot separate the Act from anything else that changed in 22 BBY, including the general state of Republic naval planning, and the ledgers do not record why any buyer signed or declined to sign.
This work extends the delivery-performance surveys of Quarn and Rennick (22 BBY) from the buyer's receipts to the yard's own ledgers, and it follows the procurement study of Pasquin (24 BBY) from the Senate's estimates to the slips on which the voted funds were spent. Oswin (23 BBY) supplied the measure of yard load by slip occupancy, and Prell (22 BBY) supplied the warning about unfinished builds that the bound in the Results answers.
References
- Pasquin, L. (24 BBY). Naval estimates and the procurement of capital ships, a Senate series. Journal of Galactic Public Finance, 14(2), 61–92.
- Quarn, D., & Rennick, A. (22 BBY). Delivery performance in capital-ship contracts, a survey from buyers' receipts. Journal of Republic Military History, 11(3), 150–174.
- Oswin, H. (23 BBY). Slip occupancy as a measure of yard load. Journal of Republic Military History, 10(3), 98–121.
- Prell, K. (22 BBY). Unfinished projects in completion-time records and the bias they leave. Proceedings of Applied Speculative Statistics, 10(1), 33–57.
- Rask, E. (21 BBY). The fleet as a fighting service, a staff note on orders and strength. Journal of Republic Military History, 12(4), 88–101.
- Delassar, C., Anthelle, V., & Kaldivar, O. (2026). The Military Creation Act of 22 BBY: A Doctrinal Analysis of Authority, Scope and the Absence of a Sunset Clause. Uncited Press.
Open in Uncited Press →