In the homogeneous equation, substitute the exponential function . The characteristic equation is
Its distinct roots are . Combining the complex-conjugate pair into real sine and cosine gives the complementary function
Five linearly independent terms supply the required five homogeneous degrees of freedom.
After integrations, the general solution is an arbitrary polynomial of degree at most :
Conversely, every such polynomial has zero th derivative. The free coefficients span the full solution space on a connected interval.
The remaining arrivals occur in the next stop phase: , or . B waits until , while A departed at , giving . This interval has length , hence
Endpoint assignments do not affect probabilities. The three outcomes are exhaustive and satisfy
These are conditional probabilities for the stated fixed ; there is no additional averaging over A's arrival phase in the question.
If B arrives during the remaining pass phase, then , or . It passes immediately, so
Because , this is strictly between zero and . The phase interval has length , giving
Condition throughout on the given arrival time of A, with . A departs at time . B also departs at exactly when it arrives before that release, so , or . Since is uniform on an interval of length ,
The original timeline below shows the three disjoint ranges of . The same idealized simultaneous-release convention as in part (a) is used.
Figure 1.
Conditional arrival ranges for the following vehicle at a fixed arrival phase of the leading vehicle
. Details.
The second moment is
Subtracting the squared expected value gives the variance and standard deviation:
The variance is nonnegative because . Both and have units of time. At the result reduces to , the standard deviation of a uniform distribution over one full cycle.
The atom at zero contributes zero to the expected value, while all positive delays must retain their unconditional density. Therefore
This mean is times the mean conditional on encountering a stop; using alone would omit the vehicles passing without delay.
The probability of a positive delay is the integral of its density:
Adding the separate no-delay event gives
This is the normalization of a probability distribution with both discrete and continuous parts.
The graph of the positive-delay probability density function is a horizontal segment at height from to , followed by zero from to . Values at individual endpoints do not affect probabilities. The separate atom at zero is not a finite-height part of this density graph.
Figure 1.
Positive-delay density and the full cumulative distribution, whose jump records the no-delay atom
. Details.
The companion cumulative distribution function in the figure is for , for , and for . Its jump at zero displays the discrete part without confusing density height with probability mass.
For a stop-phase arrival at time , the delay is . An interval of delays of width therefore corresponds to an interval of arrival phases of the same width, because . Its unconditional probability is . Hence the positive-delay part has density
This ordinary density integrates to , not to one: the remaining probability mass is at . Thus has a uniform distribution with an atom at zero, with mixing weight . Conditional on a positive delay, the density would instead be on .
The arrival phase has a uniform distribution over a cycle of length . The no-delay phases occupy length , so
Individual phase endpoints have probability zero. The model neglects clearance time and releases waiting vehicles together when the stop interval ends; the degenerate case is interpreted as release at successive cycle boundaries.
Since , substitute the proposed inverse series into the forward one. Comparing coefficients of gives
Solving successively yields the cubic series reversion coefficients
The conditions and ensure that both local series are centred at zero and that the inverse has a linear leading term.
Assume the needed derivatives exist continuously and on a neighborhood, so the inverse function theorem provides a local inverse. Put , , and . The chain rule gives , so
Differentiate once more:
Thus the first three derivatives of an inverse function are
All forward derivatives on the right are evaluated at the corresponding value .
Use . The Maclaurin series gives
All three nonzero terms are needed; stopping at fourth order would supply only two.
The derivative of the inverse tangent is . Expand it as a geometric series and integrate term by term near zero, using :
Expand the natural logarithm and reciprocal separately:
Multiplication gives the first three nonzero terms of the Maclaurin series:
The local expansions converge for .
Summing the diagonal entries gives
The eigenvalues of the inverse are the reciprocals of those of , whose sum is also . This is precisely what the trace identity predicts.

There are unlisted articles, also show them or only show them.