Вы можете попробовать следующий подход.
Предикат XPath отфильтровывает ненужные <Tool>
элементы.
SQL
-- DDL and sample data population, start
DECLARE @tbl TABLE (ID INT IDENTITY PRIMARY KEY, xmldata XML);
INSERT INTO @tbl (xmldata)
VALUES
(N'<ScoringEngine>
<Profile id="Navigation" version="1" num="4"
uniqueId="8bcf8a8b9efc4e5dad1d87510cfe6a64">
<Tool id="Payment To Income Ratio" version="1" old_id="Pmt_To_Income">
<Rule id="PaymentIncomeRatio" version="1" old_id="PTI">
<Row uniqueId="0fb11598c4224e4c97cf2afcc4e34b54" order="6"
id="0">
<Column order="1" op="RNG2" start="0.18" end="0.2"
title="Payment To Income Ratio">0.190325139</Column>
<Action name="Record Value" value="1.42085235920852"
fieldName="LO_R_PMT_TO_INCOME"/>
</Row>
</Rule>
<RecordedValue>
<Value value="1.42085235920852" fieldName="LO_R_PMT_TO_INCOME"/>
</RecordedValue>
</Tool>
<Tool id="RecentLoans" version="2">
<Rule id="RecentLoans" version="2" old_id="RecentLoans"/>
</Tool>
</Profile>
</ScoringEngine>');
-- DDL and sample data population, end
SELECT tbl.ID
, c.value('(Rule/Row/Column/text())[1]','DECIMAL(10,8)') AS [value]
FROM @tbl AS tbl
CROSS APPLY xmldata.nodes('/ScoringEngine/Profile/Tool[@id="Payment To Income Ratio"]') AS t(c);
Выход
+----+------------+
| ID | value |
+----+------------+
| 1 | 0.19032514 |
+----+------------+