Final report : April 2, 2007
Section 3 - Development of "Alternative" Source Dataset
The dataset used by the LCC model for the digital-based, or Alternative system, using the same time baseline as that used for the Baseline system, i.e., updated with FY2005 source data, is attached as the Excel file "FY2005 Calculated Values Flash All Mass Duplication April 2007.xls." It consists of an Excel workbook composed of 15 worksheets using exactly the same format for documentation and compilation of source data as that used by the file for the Baseline system, the only difference being that this workbook does not have the (last) table, "Compare 1999 to 2005."
Beginning with the workbook file developed for the Baseline system as described in Section 2, each Element was reviewed to assess what changes, if any, would be expected in an All Mass-Duplication flash memory-based system. Any such changes are summarized in the "Notes" field of the "Data Dictionary" table (the first) in Appendices B. Results from prior studies were employed in this regard, including the recently-completed DTB Distribution Study and the Machine Transition Study.
Of necessity, the Element values input to the model must reflect a system (whether the Baseline or an Alternative) in Steady-State operations. The LCC model does not provide estimates for resources used by the system other than those whose values start from given bases and are then projected into the future using one or more factors. The model, therefore, does not lend itself to the evaluation of the transition time period for the conversion of the system from an RC/CBM-based system to a DTB/DTBM-based system. Rather, it allows the comparison of two mature service delivery systems under a given set of conditions.
Therefore in order to establish the initial conditions of the Alternative system required by the LCC model, several assumptions had to be made with regard to the disposition of various assets. These assumptions are stated in the Notes field of the Data Dictionary table and also summarized below.
The tables from the 15 worksheets in the Alternative source dataset are contained in Appendices B of the report, which document the sources of all data used to generate the values of the Elements and how they are compiled. The Element numbers, names, descriptions, "FY2005" values, units of measure, and Notes are shown in the "Data Dictionary" table, which is shown first in Appendices B. Below are summarized noteworthy points regarding the data tables and data Elements.
Data Dictionary List
This table contains a listing of all the LCC model data Elements and associated values, along with the descriptions and Notes. The Source field has been hidden.
The same values for the CPI and ECI are used as those used for the Baseline system.
Readership and Circulation
The same values for readership and circulation are used as those used for the Baseline system.
The DTBM weight, packaging weight, and battery weight were all assumed to be 50% of that of the C1, while DTBM volume was assumed to be 33% of that of the C1.
The production rate of 30,000 machines is the a priori nominal production rate of DTBMs sufficient to replace total attrition (LSU and DBR), i.e., it is comparable to the rate of 42,000 CBMs per year. It is noted that the LCC model effectively doesn’t use this value, but rather calculates a steady-state replacement-level (sufficient to replace attrition) production rate for playback machines, along with the associated cost.
A unit cost of $200 per DTBM has been assumed with 2% cost inflation, based upon input from the NLS Engineering Department. The cost inflation rate for the DTBM was selected to be positive and non-zero, but lower than that of general inflation.
An MTBF of 3,000 hours has been assumed for the DTBM, which is three times the 1,000 hours of the C1. This estimate was provided by the NLS Engineering Department.
The number of Assigned Machines is assumed to be the same as that for the Baseline system, i.e., the same readers are to be served. The Machine Total Inventory is calculated as the sum of assigned, in-repair, and available machines, just as it is for the Baseline system, but it is a different value. The Machines In-Repair and Available value is calculated as the sum of in-repair and available machines, just as it is for the Baseline system, but it is a different value. The percentage is also calculated in the same manner, i.e., the sum divided by the number assigned.
The number of machines In-Repair in the Alternative system are assumed to be one-third (1/3) the number of those in the Baseline system, due to the MTBF of the DTBM being three times that of the C1.
Available machines as a percentage of assigned machines was assumed to be 10% as the starting point, which is a reasonable value and an NLS management objective. This value is considerably lower than that of the current system, wherein a (planned) surplus of CBMs exists.
The DTBM loss rate is assumed to be equal to that of the C1, which is a higher rate (6.15%) vs. one that ManTech had independently previously calculated (4.7% for all CBMs; higher for E1s and non-C1 CBMs). For consistency, the same loss rate (i.e., 6.15%) has been used for both the Baseline and Alternative systems in the LCC model.
The DTBM DBR rate is assumed to be one-third that of the C1 due to the much greater durability of the digital machine.
The total number of annual DTBM repairs in the Alternative system was estimated to be one-third that of CBM repairs in the Baseline system, because the MTBF of the DTBM is assumed to be 3 times that of the C1, based upon input from the NLS Engineering Department.
Contractor-performed DTBM repairs were assumed to be 25% of total repairs, and volunteer-performed repairs were assumed to be 75% of total repairs, based upon input from NLS.
The average (labor) cost for a contractor-performed DTBM repair was assumed to be $20, based upon input from the NLS Engineering Department.
The warranty repair rate and non-contract repair attrition rate for the DTBM were assumed to be the same values as those for C1 in the Baseline system.
DTBM disposal unit cost was assumed to be identical to that of the current system for CBMs.
Machine Usage, MTBF
This factor was assumed to be the same as that for the current system, i.e., 248.1 hours per assigned machine per year.
Machine Batteries, Spare Parts
Battery cost, battery MTBF, battery disposal cost, repair consumables (per 1,000 repairs) and repair tools (per 1,000 repairs) for the Alternative system were assumed to be the same as the corresponding values for the Baseline system.
The LCC model requires an Element input which is the percentage of the cost of a new machine required in parts to perform the average repair. For the DTBM this is 7.5% ($15/$200) versus 5% (very nearly $14/$280) for the C1.
The Element values of the Alternative system for mastering unit cost, new container cost (assumed to be the same as for the 4-cassette container), titles produced per year, additional mastering cost, average length, and average copies per title were assumed to be identical to those for the Baseline system.
The number of book media units per copy will be 1.0 for the Alternative system versus 2.2 for the Baseline system. NLS is planning to use 128, 256, and 512 MB cartridges in the future system as appropriate for book titles of varying lengths.
There would be no additional duplication costs (associated with Intermasters) for the Alternative system as there is for the Baseline system.
The 4-cassette container and DTB container are assumed to be the same unit cost when purchased new. The same numbers of containers are assumed to be reconditioned and scrapped in both systems, and at the same unit costs to process (recondition or inspect/dispose).
The book media unit cost includes a weighted-average unit cost for both new and reconditioned flash memory cartridges. This cartridge average unit cost is $3.13, which assumes a 1% disposal rate for weeded/inspected cartridges so that 49.5% of DTB copies produced will use reconditioned cartridges, and new cartridges will cost $6 each on average. The unit cost to recondition a cartridge is assumed to be $0.20.
A duplication unit cost of $0.88 is estimated, which includes the actual duplication process (loading DTB files onto an NLS cartridge), cartridge labels and labelling, container labels and labelling, and loading the cartridge into the container. Thus a total cost of $4.01 per DTB copy is estimated for steady-state operations (which includes all book production costs except those for narration and the container itself).
Magazines and TBT
The characteristics and costs of the magazine program were assumed to remain unchanged. ManTech had no basis for estimating the costs of the magazine/TBT program for any system other than that of the Baseline, i.e., RC-based, because magazine operations were not addressed in the DTB Distribution Study.
Production Inflation Rates
Both the mastering and duplication inflation rates for the Alternative system were assumed to be the same as those for the Baseline system. There is no basis for making any change in the mastering inflation rate, and little is currently known about the DTB mass-duplication process.
Network & MSC Figures
The Data Dictionary table, Notes field, details which network agency or MSC costs in the Alternative system are expected to vary from those of the Baseline system, the expected magnitude of the variations, and the reasons for the variations.
It has been assumed, based upon NLS input, that MSCs will not handle DTBs in any manner, either for archival storage or for circulating and/or backup collections, nor will there be any duplication of DTBs in MSCs.
It was assumed that all machine and supply operations in MSCs, and their associated costs, will be the same in the Alternative system as they are in the Baseline system.
Facilities Inflation Rates
The facilities inflation rate for the Alternative system was assumed to be the same as that for the Baseline system.
The USPS costs for the Alternative system were assumed to be the same as those for the Baseline system. Almost exactly the same number of book containers would be handled by the USPS under both systems on a two-way delivery basis. While there would be some reduced machine deliveries in the Alternative system, this difference in workload would be insignificant compared to the total workload associated with books. While the DTB container will be slightly smaller than the RC container and thus there would be some reduction in USPS transportation costs, the same amount of sorting and handling would be necessary under both systems.